What Demands Attention in Urban Street Scenes? From Scene Understanding towards Road Safety: A Survey of Vision-driven Datasets and Studies

Julie Stephany Berrio; Mao Shan; Stewart Worrall; Yaoqi Huang

REVIEW 5 major objections 3 minor 247 references

What Demands Attention in Urban Street Scenes? From Scene Understanding towards Road Safety: A Survey of Vision-driven Datasets and Studies

T0 review · 5 major / 3 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This survey claims that all road-safety-relevant scene-understanding research can be organized by a two-group, eleven-category taxonomy of attention-worthy elements, and uses it to analyze 40 vision-driven tasks and 78 datasets.

desk verdict A genuinely useful dataset survey with a sensible taxonomy, but the authors need to clean up their own scoring: headline numbers disagree with the abstract, and a few taxonomy assignments contradict their own definitions. read the letter →

arxiv 2507.06513 v3 pith:C75BYUHC submitted 2025-07-09 cs.CV

Yaoqi Huang , Julie Stephany Berrio , Mao Shan , Stewart Worrall This is my paper

classification cs.CV

keywords computervisionsceneunderstandingroadsafetytrafficanomalydetectiondatasetsurveytaxonomyautonomousdrivingbenchmarkanalysis

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This survey sets out to show that the many vision tasks used for urban road safety are not a scattered collection but can be ordered by one question: what element in the scene demands attention and why. It proposes a taxonomy with two main groups—anomalies (abnormal entities) and pertinent entities (normal but critical elements)—spanning eleven categories and twenty-three research topics. On this basis it analyzes 40 vision-driven tasks and 78 datasets, and claims that its functional classification reveals which tasks lack public benchmarks and which datasets have flawed or mismatched annotations. A sympathetic reader would take the paper's central offer to be a unified analytical framework that lets researchers see connections among historically isolated fields and choose datasets strategically.

What carries the argument

The load-bearing mechanism is the taxonomy itself, organized around the question 'Which elements are most critical in the current traffic scenario and require attention to road safety?' It is a two-level hierarchy: the reason the element demands attention (anomaly versus pertinent entity) and the specific category of that reason (spatial location, semantic category, event, kinematic pattern, status or condition, appearance, or multiple reasons for anomalies; spatial location, semantic category, status or condition, or kinematic pattern for pertinent entities). Each leaf node in the taxonomy's graphical diagram maps to the corresponding tasks and representative datasets, with gray task nodes marking the absence of publicly available datasets. The taxonomy does the work of the paper: it is the standard against which tasks are assigned, datasets are reclassified, and research gaps are identified.

What would settle it

Re-examine a sample of the reclassified datasets at the annotation level: if Pothole-600's pixel masks are not genuinely segmentation-grade, if GLARE frequently annotates more than one traffic sign per image, or if a newly published vision-driven road-safety dataset cannot be assigned to any of the eleven categories without stretching their definitions, the paper's reclassification claims and its claim of comprehensive coverage would be called into question.

Watch

Extended reading notes

Core claim

The paper's central claim is that 'critical traffic elements that demand attention' can be classified by their functional role in road safety, and that this classification fully organizes the field. The resulting taxonomy has two groups: anomalies, which demand attention because they are abnormal (by spatial location, semantic category, event, kinematic pattern, status or condition, appearance, or a hybrid of these), and pertinent entities, which are normal but critical to the current driving maneuver (by spatial location, semantic category, status or condition, or kinematic pattern). The paper reports that this yields eleven categories and twenty-three common research topics, and it uses the taxonomy to examine 40 vision-driven tasks and 78 datasets, reclassifying each dataset according to what its data and annotations actually support rather than its stated task. It further claims that cross-domain investigation reveals substantial variations in benchmark quality, with recurring limitations such as uneven task coverage, imbalanced distributions, inconsistent or insufficient annotations, and limited multimodal and cross-task support.

Load-bearing premise

The survey's conclusions about coverage and benchmark quality rest on the authors' unpublished qualitative reclassification of each dataset and on their manual visual inspection of annotation files, so if those judgments are inconsistent or incomplete, the taxonomy's assignments and the research-gap findings would change.

Editorial extensions

If this is right

Researchers can use the taxonomy to position new tasks and datasets relative to the full field, making it easier to transfer methods between related but historically isolated areas such as obstacle segmentation, anomaly segmentation, and accident anticipation.
The functional classification exposes concrete gaps: road construction detection, traffic salient object detection, and appearance-based anomaly tasks (criminal recognition, suspicious vehicle recognition, damaged vehicle detection) currently lack public datasets.
Datasets with mismatched claimed tasks are reclassified by what their annotations actually support, so benchmarks such as Pothole-600 become segmentation resources rather than detection resources, and GLARE is treated as single-sign traffic sign detection.
The distinction between anomaly and pertinent entity changes how accidents are categorized: ego-involved collisions are treated as anomalous kinematic patterns, while accidents witnessed by the ego vehicle are treated as anomalous events.
The survey's cross-domain analysis supports a push toward unified annotation standards and dataset reuse, because many datasets can serve multiple categories once their functional format is recognized.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the taxonomy is adopted by the community, a natural next step would be a unified benchmark suite that samples one or more datasets from each of the eleven categories, allowing holistic road-safety perception to be measured instead of task-isolated accuracy.
The authors' practice of reclassifying datasets by annotation format implies that published task names are not a reliable guide to dataset utility; a testable extension would be requiring dataset authors to report the functional annotation level (classification, detection, segmentation, graph, temporal) as metadata.
The taxonomy's functional-role principle could be carried beyond vision into multimodal and planning-oriented settings, since the same attention-worthy elements appear in LiDAR point clouds and in prediction or decision-support pipelines that the survey deliberately excludes.
A direct editorial check would be to compute inter-annotator agreement on assigning a sample of datasets to the eleven categories; low agreement would signal that the taxonomy needs sharper definitions, while high agreement would support its claim to be a stable organizing framework.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The short version: this is a useful survey of traffic-scene perception datasets, organized around a plausible taxonomy of 'what demands attention' — anomalies versus pertinent entities. It earns its keep through the concrete dataset analysis: the authors actually looked at annotation files and name specific flaws (dummy boxes in CADP, label mismatches in CST-S3D, dead links in A3D). That hands-on detail is what makes a survey a reference rather than a filing cabinet.

What is new is the two-group, eleven-category taxonomy that bridges previously siloed tasks (accident anticipation, lane detection, road damage, etc.). It is not a conceptual breakthrough—it is a reorganization—but it is an honest and mostly sensible one, and the coverage of 40 tasks and 78 datasets with visualizations represents a lot of work.

Now the soft spots, in proportion. First, the numbers in the arXiv abstract (10 categories, 20 subclasses, 35 tasks, 73 datasets) do not match the full text (11, 23, 40, 78). For a survey whose whole purpose is organized coverage, that is sloppy and needs fixing before anyone cites it as authoritative.

Second, and more substantive: the taxonomy assignments are not always reproducible from the stated rule. GLARE is filed under Traffic Sign Detection even though the text says it localizes only one sign per image—by the paper's own Section 2.2/2.3 distinction, that is localization, not detection. Driver Attention Prediction sits under Kinematic Pattern, but attention maps are not kinematic patterns of traffic elements. RiskBench under 'Multiple Reasons' is more defensible, since the dataset as a whole covers multiple anomaly types even if each sample has one, but the other two look like narrative convenience. The reclassification protocol is not released, so readers cannot verify the assignments or the 'research gap' claims built on them. This does not sink the survey—most assignments look right, and the categories are clear enough—but it is a real quality issue for a paper that aims to be a standard reference.

Who this is for: anyone working on road-safety perception who needs to choose datasets or understand the field map. It deserves a serious referee; the work is substantial and the issues are correctable.

Recommendation: engage with it, but require the authors to align the abstract with the full text, audit category assignments against their own definitions, and release a classification table with one-line justifications for non-obvious cases.

Referee Report

5 major / 3 minor

Summary. This survey proposes a taxonomy of attention-worthy traffic elements for road safety, organizing them into two main groups (anomalies and pertinent entities), eleven categories, and twenty-three research topics, and uses it to review 40 vision-driven tasks and 78 datasets. The paper provides detailed descriptions, annotation visualizations, and per-dataset quality critiques, and it draws cross-domain conclusions about research gaps and benchmark limitations.

Significance. If the taxonomy is consistent and the dataset assignments are sound, this survey would provide a valuable unified framework for a fragmented field, with practical utility for dataset selection and gap identification. The paper's concrete annotation visualizations and specific, checkable claims about dataset flaws (e.g., dummy boxes in CADP, label mismatches in CST-S3D, broken links in A3D) are strengths that go beyond typical survey-level description. The central claim, however, depends on the correctness and transparency of the qualitative reclassification that underlies the taxonomy assignments, and that process is neither fully specified nor reproducible, which is the main risk to the paper's contribution.

major comments (5)

[Section 1 and Abstract] The inventory counts are inconsistent between the abstract (10 categories, 20 subclasses, 35 tasks, 73 datasets) and the full text (11 categories, 23 subclasses, 40 tasks, 78 datasets). Since the survey's contribution is framed as a complete organization of the field, these mismatches must be reconciled and a single set of counts used consistently throughout the manuscript and all tables.
[Section 3.9.7] The text states that GLARE actually localizes only one traffic sign per image, yet GLARE is placed under Traffic Sign Detection. By the paper's own definitions in Sections 2.2 and 2.3, single-instance localization is object localization, not detection. This is an internal inconsistency in applying the stated functional-format rule, and it affects the taxonomy's credibility for this category.
[Section 3.7.1] RiskBench is the sole dataset in the category 'Anomaly Due to Multiple Reasons' / Risk Identification, but the text admits that it 'exclusively labels one predefined type of risks per sample.' This placement contradicts the category's own justification, which requires multi-factor anomalies. The assignment should be reconsidered, or the category definition and the dataset description should be aligned.
[Section 3.4.3] Driver Attention Prediction is filed under 'Anomaly Due to Kinematic Pattern', yet the attention maps in BDD-A and DADA-2000 are derived from gaze and fixation patterns, not from kinematic patterns of traffic elements. This is a categorical mismatch with the taxonomy's stated reason for the category, and it undermines the claim that category membership is determined by the functional-format criterion.
[Section 3 and Table 2] The dataset-to-taxonomy assignments rest on an unpublished, qualitative reclassification of each dataset (e.g., Pothole-600 as segmentation, GLARE as detection, UCF-Crime as partially traffic-related), with no protocol and no released assignment table. Because the 'research gap' conclusions depend directly on these assignments, the paper should specify the reclassification procedure and provide a complete, citable appendix of assignments so that the claimed coverage can be independently verified.

minor comments (3)

[Abstract] There is a missing space between 'Comparedto' and 'existingsurveys' in the abstract; these and other spacing errors should be corrected during copyediting.
[Section 3.3.4] The paper notes that all 204 YouTube links in A3D are inaccessible, which prevents data retrieval and verification. This should be flagged more prominently in Table 2 (e.g., with a data-availability marker) so that readers are not misled about the dataset's current usability.
[Table 2] The table's caption defines 'Level' abbreviations (C, L, D, S, I, P), but the table itself does not indicate which granularity of label is used for 'D' in datasets where detection is only claimed by the original authors; adding a footnote that distinguishes claimed vs. reclassified labels would improve clarity.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the taxonomy is a definitional construct; insider datasets (STU, M2S-RoAD) and inconsistent assignments (GLARE, RiskBench) are reproducibility concerns, not reductions to inputs.

full rationale

The paper's load-bearing claim is that its taxonomy, namely two groups, eleven categories, twenty-three topics, 40 tasks, and 78 datasets, organizes the field of vision-driven road-safety perception. That claim is a definitional construct, not a derived result: the paper states the taxonomy 'was initially derived from thirteen distinct perspectives, integrating insights from traffic scene understanding literature and practical road-safety considerations' and then grouped by whether the element is abnormal or normal-but-critical. There is no equation, no fitted parameter, and no predicted quantity that reduces to an input, so no classic circularity pattern (self-definitional derivation, fitted input called prediction, or uniqueness imported from prior work) is present. The survey is self-contained against external benchmarks: nearly all 78 datasets are external resources, and the taxonomy structure would stand even if every dataset were replaced. Two entries appear to come from the authors' own research group (STU [47], M2S-RoAD [64]), but neither is load-bearing: the Anomaly Segmentation category contains five independent 2D benchmarks alongside STU, and Road Damage Segmentation contains PotholeMix alongside M2S-RoAD, so the central organizing claim does not depend on these entries. The concerns worth flagging are reproducibility and consistency, not circularity. The stated rule that each dataset is classified 'according to its actual functional format' is applied inconsistently: GLARE is kept under Traffic Sign Detection although the paper says it 'actually only localizes one traffic sign per image,' which the paper's own Section 2.2/2.3 single-instance vs. multi-instance distinction places in object localization; RiskBench is the sole occupant of 'Anomaly Due to Multiple Reasons' although the paper admits it 'exclusively labels one predefined type of risks per sample.' The Road Construction Detection gap partly results from the paper's own scoping exclusion of construction-site datasets ('exhibit a clear domain shift from our focus on urban street scenes and are therefore not further analyzed'), and the abstract (10 categories, 35 tasks, 73 datasets) contradicts the full text (11 categories, 40 tasks, 78 datasets). These are correctness and verifiability risks for the survey's empirical claims, to be weighed in referee review, not evidence that the taxonomy reduces to its inputs.

Assumptions & free parameters 0 free parameters · 2 assumptions · 0 invented entities

This is a survey, so the central claim rests on no fitted parameters. The load-bearing assumptions are the representativeness of the surveyed literature and the validity of the authors' qualitative dataset classification; both are stated but not mechanically operationalized.

assumptions (2)

domain assumption The 78 surveyed datasets and 40 tasks are representative of the field of vision-driven road-safety scene understanding.
The dataset selection relied on major academic publication platforms and reference tracing (Section 1, Scope), but no systematic meta-analysis protocol or exhaustive search log is provided, so completeness is asserted rather than demonstrated.
domain assumption The authors' manual reclassification of each dataset to its 'actual functional format' is faithful and consistent.
Section 1 states that 'each surveyed dataset is classified according to the type of attention-worthy traffic element it addresses and the form in which that element is represented,' and concrete reclassifications (e.g., Pothole-600 as segmentation) rest on visual inspection of public packages, which is a subjective judgment.

how reviews work

0 comments

Cite this review

Pith. "Pith review of What Demands Attention in Urban Street Scenes? From Scene Understanding towards Road Safety: A Survey of Vision-driven Datasets and Studies." pith.science (2026). https://pith.science/paper/C75BYUHC

@misc{pith2026250706513,
  author       = {Pith},
  title        = {Pith review of: What Demands Attention in Urban Street Scenes? From Scene Understanding towards Road Safety: A Survey of Vision-driven Datasets and Studies},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/C75BYUHC}},
  note         = {Machine review of arXiv:2507.06513}
}

read the original abstract

Advances in vision-based sensors and computer vision algorithms have significantly improved the analysis and understanding of traffic scenarios. To facilitate the use of these improvements for road safety, this survey systematically categorizes the critical elements that demand attention in traffic scenarios and comprehensively analyzes available vision-driven tasks and datasets. Compared to existing surveys that focus on isolated domains, our taxonomy categorizes attention-worthy traffic entities into two main groups that are anomalies and normal but critical entities, integrating ten categories and twenty subclasses. It establishes connections between inherently related fields and provides a unified analytical framework. Our survey highlights the analysis of 35 vision-driven tasks and comprehensive examinations and visualizations of 73 available datasets based on the proposed taxonomy. The cross-domain investigation covers the pros and cons of each benchmark with the aim of providing information on standards unification and resource optimization. Our article concludes with a systematic discussion of the existing weaknesses, underlining the potential effects and promising solutions from various perspectives. The integrated taxonomy, comprehensive analysis, and recapitulatory tables serve as valuable contributions to this rapidly evolving field by providing researchers with a holistic overview, guiding strategic resource selection, and highlighting critical research gaps.

Figures

Figures reproduced from arXiv: 2507.06513 by the authors.

**Figure 1.** Graphical taxonomy diagram of attention-worthy elements for road safety, organized by reason (anomaly vs. pertinent entity) and category. Leaf nodes list the corresponding tasks and representative datasets. Categories with empty task node indicate the lack of well-discussed tasks. Tasks shown in gray indicate a lack of publicly available datasets. focus on individuals or groups with abnormal behavior [11, 13, 14, 15… view at source ↗

**Figure 2.** Publication statistics of 78 surveyed datasets by publication venue (left) and year (right). CVPR (IEEE/CVF Conf. on Computer Vision and Pattern Recognition), ICCV (IEEE/CVF Int. Conf. on Computer Vision), ECCV (European Conf. on Computer Vision), ACM MM (ACM Int. Conf. on Multimedia), ICCVW/CVPRW (ICCV/CVPR Workshops), ACCV (Asian Conf. on Computer Vision), AVSS (IEEE Int. Conf. on Advanced Video and Signal-based S… view at source ↗

**Figure 3.** Representative examples from Obstacle Segmentation or Anomaly Segmentation datasets, showcasing raw images with their corresponding ground truths (extracted from image or JSON files and overlaid on the originals for demonstration). For semantic masks, when RGB ground-truth annotations are available, the original color scheme is preserved with modified alpha values to enhance visibility. Otherwise, high-contrast colo… view at source ↗

Figures from the paper (57 more)

**Figure 4.** Figure 4: Sample images overlaid by labels in traffic congestion classification datasets [48, 49]. Colors were randomly assigned to enhance visual differentiation among classes. entry and exit of large vehicles and equipment and introduce additional obstacles and narrow lanes, l…

**Figure 5.** Figure 5: TADD [50] illustration with category label highlighted. (a) Irrelevant (b) Non-vehicular (c) Near-miss Incidents (d) Sequential Sample Frames from Incorrectly Concatenated Segment (e) Dummy Spatial and Temporal Annotations Extracted from JSON file [PITH_FULL_IMAGE:fig…

**Figure 6.** Figure 6: CADP[20] illustration: issues in footage and labels. (d), extracted from the same segment, a significant portion of CADP inappropriately mixes multiple irrelevant video sequences. Moreover, although temporal and spatial labels are created at the start and end frames fo…

**Figure 7.** Figure 7: Sample images of CCD [21] with labels extracted from the TXT file and interpreted to enhance readability. We overlay red warning texts and bottom process bars for demonstration. participants, as clarified in [190] and demonstrated in [PITH_FULL_IMAGE:figures/full_fig_…

**Figure 8.** Figure 8: Sample frames of DAD [13]. For better readability, instance-level labels are just overlaid for incident participants. accident scam (e.g. training clip 354). Meanwhile, frame-based labels for moving objects, including the bounding box, semantic category, tracking ID, a…

**Figure 9.** Figure 9: DoTA [11] Illustration: What-Where-When pipeline. (A) Cause Frame (B) Effect Frame [PITH_FULL_IMAGE:figures/full_fig_p022_9.png]

**Figure 10.** Figure 10: Each column shows the sample frames of one accident clip in CTA [51], with temporal and categorical labels. with abnormal behavior already rushes out in the 52nd frame (top-middle) and continues its motion in the 71st frame (bottom-right). Moreover, spatial annotation…

**Figure 11.** Figure 11: Illustration of multi-level labels in MM-AU [52]. (e.g., rollover), respectively. It selects 1935 accident videos from YouTube channels, predominantly at resolutions of 1280×720 and 640×360. Instead of providing pre-extracted clips, CTA releases video IDs and temporal…

**Figure 12.** Figure 12: Sample traffic crime frames extracted from [53, 29], overlaid with categorical and temporal (if available) labels. ‘motorcycle’ (top-right images). In the bottom images, the bridge is incorrectly labeled as ‘car’, while the partial car is mislabeled as ‘truck’. In add…

**Figure 13.** Figure 13: CST-S3D [55] illustration with labels overlaid. category, they focus on ‘intentions’ and have a large proportion of proper behavior footage. Hence, we analyze them in Section 3.8. YouTubeCrash: YouTubeCrash [14] collects dashcam YouTube videos to evaluate dangerous ve…

**Figure 14.** Figure 14: Sample frames extracted from YouTubeCrash and GTACrash [14] with labels overlaid for demonstration. BDDA [8] DADA-2000 [9] [PITH_FULL_IMAGE:figures/full_fig_p027_14.png]

**Figure 15.** Figure 15: Sample frames of driver attention prediction datasets [8, 9], overlaid by attention maps for demonstration. 3.4.3. Driver Attention Prediction The driver attention prediction aims to identify and locate regions or objects in the driving environment that are likely to …

**Figure 16.** Figure 16: Sample images and their corresponding ground-truth binary maps from Crack Segmentation datasets. been resized to 480×320. Manual pixel-level annotations are available only for the first 118 images in MAT format, which further limits CFD as a standalone supervision sou…

**Figure 17.** Figure 17: Pothole-600 [61] illustration: image and annotations [PITH_FULL_IMAGE:figures/full_fig_p030_17.png]

**Figure 18.** Figure 18: CQU-BPDD [62] illustration: images and class labels. delineation of pothole regions rather than simple detection. As illustrated in [PITH_FULL_IMAGE:figures/full_fig_p030_18.png]

**Figure 19.** Figure 19: PotholeMix [63] illustration. Sample image (left) and corresponding mask (second from left) from the subset [200] in image collection; Sample image (middle), mask (second from right), and disparity map (right) in video collection [PITH_FULL_IMAGE:figures/full_fig_p03…

**Figure 20.** Figure 20: M2S-RoAD [64] illustration: images and class labels [PITH_FULL_IMAGE:figures/full_fig_p031_20.png]

**Figure 21.** Figure 21: GAPs [65] illustration (from left to right): crack, pothole, invalid patch, applied patch, and open joints. were manually created by the authors. As shown in [PITH_FULL_IMAGE:figures/full_fig_p031_21.png]

**Figure 22.** Figure 22: RDD2022 [66] images with spatial and class labels. GAPs: Originally introduced in 2017, GAPs [177] contains 1969 grayscale images collected from three federal roads in Germany and was extended to 2468 images (1920×1080) in 2019 with additional images from a new federa…

**Figure 23.** Figure 23: Sample images in four adverse conditions from ACDC [67], collected in urban and rural areas. (a) Sample images and corresponding semantic annotations from NightDriving [71] (left two) and NightCity [72] (right two). (b) DarkZurich [73]: Sample Nighttime Image (left), …

**Figure 24.** Figure 24: Illustration of nighttime datasets [71, 72, 73] 3.5.7. Illumination Perception and Adaption Accurate perception of illumination allows automatic activation of night mode without driver intervention and adaptation of headlight beam patterns based on road conditions, co…

**Figure 25.** Figure 25: Sample risky frame in RiskBench [67], with the ground-truth annotation of the corresponding risk. 3.7. Anomaly Due to Multiple Reasons During a journey, the ego vehicle may encounter any of the aforementioned anomalies that can jointly affect road safety. Accordingly,…

**Figure 26.** Figure 26: Sample frames in pedestrian intention prediction datasets [15, 77], with multidimensional labels overlaid. boxes for non-crossing pedestrians. However, the official dataset request link has become unavailable due to technical and legal issues, restricting accessibilit…

**Figure 27.** Figure 27: Sample frame with annotations of STIP [78], captured by left, center, and right front-mounted cameras [PITH_FULL_IMAGE:figures/full_fig_p038_27.png]

**Figure 28.** Figure 28: Sample annotated frame from [79], captured using calibrated and time-synchronized stereo cameras and LiDAR. varying illumination conditions, frequent occlusions, diverse pedestrian appearances, and complex background contexts, have resulted in a comparative scarcity o…

**Figure 29.** Figure 29: Sample frames from ECPDP [81]. The overlaid labels include bounding boxes for all instances, joint coordinates for pedestrians and riders, and orientations for eligible persons. For better illustration, category and occlusion labels for ‘pedestrian’ (green) and ‘motor…

**Figure 30.** Figure 30: Sample frames from two 1043-syn videos [82] reported as ‘risky’ and ‘safe’, with main labels overlaid [PITH_FULL_IMAGE:figures/full_fig_p040_30.png]

**Figure 31.** Figure 31: Sample frames in sub-NUDrive [82] reveal that multiple lane-change maneuvers contained in the single clip. CARLA-syn: Two synthetic lane-changing video datasets, 271-syn and 1043-syn [82], are generated using CARLA [217] at a resolution of 1280×720. Spatial relations …

**Figure 32.** Figure 32: Sample frames in Caltech [85] with category label and instance ID overlaid for each bounding box. The top three rows crop the lower image regions to better visualize distant objects, while the last row presents uncropped images. CityPersons: CityPersons [85] augments …

**Figure 33.** Figure 33: Illustration of pedestrian detection datasets [74, 85] with bounding boxes overlaid on sample frames [PITH_FULL_IMAGE:figures/full_fig_p043_33.png]

**Figure 34.** Figure 34: Sample frames in ECP [86]. The overlaid annotations include bounding boxes for instances in all six classes and orientation labels for pedestrians and vehicles of riders. an occlusion level. As shown in [PITH_FULL_IMAGE:figures/full_fig_p043_34.png]

**Figure 35.** Figure 35: Representative visible and infrared images in LLVIP dataset [87], with annotation overlaid for ‘person’ instances. despite being closer to and more relevant to the ego vehicle than distant pedestrians on the sidewalk who are labeled as ‘pedestrian’. This inconsistency…

**Figure 36.** Figure 36: Sample images in TDC [88], with bounding boxes and instance ids overlaid, where green, cyan, and yellow boxes denote cyclists, pedestrians, and other riders, respectively. release contains 15,488 image pairs at 1280×1024 pixels. Second, LLVIP records pedestrian behavi…

**Figure 37.** Figure 37: Sample frames with annotations in TuSimple [89]. dedicated negative subset [88], as shown in the second scenario in [PITH_FULL_IMAGE:figures/full_fig_p046_37.png]

**Figure 38.** Figure 38: Sample frames in CULane [90], overlaid with lane annotations. Colors are assigned for illustration. smooth and are even not aligned with the actual lane markings, causing models to learn inaccurate lane geometry. Furthermore, there are several inconsistencies in the a…

**Figure 39.** Figure 39: Examples in the sample package [227] of BDD100K [42], overlaid with lane detection annotations. Colors are assigned to distinguish lane markings with different attributes. based on the sample package provided by [227]. As illustrated in [PITH_FULL_IMAGE:figures/full_…

**Figure 40.** Figure 40: Sample frames from VIL-100 [91] with corresponding segmentation maps and JSON-derived annotations [PITH_FULL_IMAGE:figures/full_fig_p049_40.png]

**Figure 41.** Figure 41: Sample frame with 2D lane detection annotations from OpenLane-V [94]. Colors are assigned for illustration. VIL-100: As the first video-based lane dataset [94] introduced in 2021, VIL-100 [91] aims to extend lane detection from single images to video sequences and use…

**Figure 42.** Figure 42: Sample frames and corresponding semantic annotations of ApolloScape [93] provided on the repository [229] [PITH_FULL_IMAGE:figures/full_fig_p050_42.png]

**Figure 44.** Figure 44: Sample frames from nuScenes [95] with corresponding HD map annotations or 3D object annotations [PITH_FULL_IMAGE:figures/full_fig_p052_44.png]

**Figure 45.** Figure 45: Surround-view images with corresponding HD map annotations (yellow: lane divider, green: road boundary, blue: crosswalk) of sample scenes in Argoverse2 [96]. for camera-based perception such as (b) in [PITH_FULL_IMAGE:figures/full_fig_p052_45.png]

**Figure 46.** Figure 46: Sample images in DFG [97]. The overlays include bounding boxes, semi-transparent masks, and category labels. 3.9.7. Traffic Sign Detection Compared to the traffic sign recognition task that classifies single-sign images, traffic sign detection localizes all traffic si…

**Figure 47.** Figure 47: Illustrations of images and labels from TT100K [98] particularly suitable for specialized small-object detection algorithms. As shown in the top-right image of [PITH_FULL_IMAGE:figures/full_fig_p054_47.png]

**Figure 48.** Figure 48: GTSDB [99] illustration: (A) Raw images with bounding boxes and category labels. Repository-provided (B) ‘semantic’ and (C) ‘instance’ masks, though all instances use identical pixel values. Instance contours in (C) are highlighted. additional panels, and other signs.…

**Figure 49.** Figure 49: Sample images in GLARE [75], overlaid with bounding box and category label for annotated traffic sign. 3.9.8. Traffic Sign Graph Generation The comprehensive understanding of traffic signs extends beyond instance-level localization and classification to include the re…

**Figure 50.** Figure 50: Sample Images in CTSU [100], showing sign components with overlaid bounding boxes, instance IDs, and category labels, and representing relationships with overlaid arrows. evaluation reliability, since apparent prediction errors may actually reflect annotation mistakes…

**Figure 51.** Figure 51: RS10K [101] illustration: (A) images with overlaid entity polygons and labels, including road (yellow), lane (green), sign (cyan), and sign content that contain texts (red), arrows (pink), and symbols (orange); (B) visualization of relation annotations that further in…

**Figure 52.** Figure 52: Cropped samples in LISA [102], overlaying image tag and spatial and category labels for annotated traffic lights. (a) Examples in ‘Green’ and ‘GreenLeft’ Classes (b) Max Density Scenario and Unsignalized intersection (c) Inconsistent Labels for the Same Light in Conse…

**Figure 53.** Figure 53: Cropped sample images in BSTLD [103], overlaying spatial and category labels for annotated traffic lights. vehicle lights, especially during precipitation, and d) regional diversity in orientation (e.g. vertical), signal types (e.g. arrow), and supplementary displays …

**Figure 54.** Figure 54: Sample Images in DriveU [104], overlaying frame-level and object-level labels. Colors of bounding boxes represent ‘state’ attribute. Texts beneath each bounding box denote (from top to bottom) ‘unique ID’, ‘pictogram’, and ‘relevance’. and average in the test set are …

**Figure 55.** Figure 55: Sample images (cropped) in S2TLD [105], overlaying spatial and categorical labels for annotated traffic lights. proportion of unknown (27.2%) and off (7.5%) may dilute the supervisory value of finer-grained categories, biasing algorithms toward less discriminative cat…

**Figure 56.** Figure 56: Sample frames cropped from TPGR videos [106], overlaying classification label and its interpretation. sample frames extracted from TPGR videos. Among the texts overlaid on the images, only the numeric class ID comes from the annotation files. The video ID and frame ID…

**Figure 57.** Figure 57: Sample frame from BLVD [108], showing synchronized RGB imagery and 3D point clouds, with 3D bounding boxes for each traffic participant and the corresponding 5D interactive maneuver type annotated along each trajectory. resolution of 1920×500 pixels, a 64-beam LiDAR w…

**Figure 58.** Figure 58: Distribution of the 78 surveyed datasets by data source, perspective, scenario type, and sensor type. 4.1.3. Data Sources As summarized in [PITH_FULL_IMAGE:figures/full_fig_p067_58.png]

**Figure 59.** Figure 59: Country-level data collection distribution across 59 datasets (106 country entries total), grouped by region. Cities per dataset Datasets % of 78 % of 45 1 20 25.6% 44.4% 2–5 11 14.1% 24.4% 6–10 8 10.3% 17.8% 11+ 6 7.7% 13.3% Total (specified) 45 57.7% 100% Unspecifie…

**Figure 60.** Figure 60: Distribution of datasets by number of cities covered. % of 78 is relative to all surveyed datasets (■ bars scaled to 𝑛 = 33, unspecified); % of 45 is relative to datasets with a city count (■ bars scaled to 𝑛 = 20, single-city datasets). be generalized to the deployme…

**Figure 61.** Figure 61: Distribution of 78 surveyed datasets by annotation task: Solid bars for datasets exclusively supporting that task; Faded bars for datasets that also support other tasks; The combined length represents the total amount for that task. Task-annotation Misclassification: …

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

247 extracted references · 32 canonical work pages

[1]

Road traffic injuries,https://www.who.int/news-room/fact-sheets/detail/road-traffic-injuries (2023)

2023
[2]

B. Tian, B. T. Morris, M. Tang, Y. Liu, Y. Yao, C. Gou, D. Shen, S. Tang, Hierarchical and networked vehicle surveillance in its: A survey, IEEE Transactions on Intelligent Transportation Systems 18 (1) (2017) 25–48.doi:10.1109/TITS.2016.2552778

arXiv 2017
[3]

9396–9405.doi:10.1109/CVPR.2019.00963

A.Kirillov,K.He,R.Girshick,C.Rother,P.Dollár,Panopticsegmentation,in:2019IEEE/CVFConferenceonComputerVisionandPattern Recognition (CVPR), 2019, pp. 9396–9405.doi:10.1109/CVPR.2019.00963

arXiv 2019
[4]

Y.Zhou,Y.Zhang,Z.Zhao,K.Zhang,C.Gou,Towarddrivingsceneunderstanding:Aparadigmandbenchmarkdatasetforego-centrictraffic scene graph representation, IEEE Journal of Radio Frequency Identification 6 (2022) 962–967.doi:10.1109/JRFID.2022.3207017

arXiv 2022
[5]

Huang, X

Y. Huang, X. Wang, Hazards prioritization with cognitive attention maps for supporting driving decision-making, IEEE Transactions on Intelligent Transportation Systems 25 (11) (2024) 16221–16234.doi:10.1109/TITS.2024.3413675

arXiv 2024
[6]

L. Qin, Y. Shi, Y. He, J. Zhang, X. Zhang, Y. Li, T. Deng, H. Yan, Id-yolo: Real-time salient object detection based on the driver’s fixation region, IEEE Transactions on Intelligent Transportation Systems 23 (9) (2022) 15898–15908.doi:10.1109/TITS.2022.3146271

arXiv 2022
[7]

N. Jia, Y. Sun, X. Liu, Tfgnet: Traffic salient object detection using a feature deep interaction and guidance fusion, IEEE Transactions on Intelligent Transportation Systems 25 (3) (2024) 3020–3030.doi:10.1109/TITS.2023.3293822

arXiv 2024
[8]

Y. Xia, D. Zhang, J. Kim, K. Nakayama, K. Zipser, D. Whitney, Predicting driver attention in critical situations, in: Asian Conference on Computer Vision, 2017. URL https://api.semanticscholar.org/CorpusID:52019549

2017

Show all 247 references

[9]

J. Fang, D. Yan, J. Qiao, J. Xue, H. Yu, Dada: Driver attention prediction in driving accident scenarios, IEEE Transactions on Intelligent Transportation Systems 23 (6) (2022) 4959–4971.doi:10.1109/TITS.2020.3044678

2022
[10]

K. K. Santhosh, D. P. Dogra, P. P. Roy, Anomaly detection in road traffic using visual surveillance: A survey, ACM Comput. Surv. 53 (6) (Dec. 2020). doi:10.1145/3417989. URL https://doi.org/10.1145/3417989

2020 doi
[11]

Y. Yao, X. Wang, M. Xu, Z. Pu, Y. Wang, E. Atkins, D. J. Crandall, DoTA: Unsupervised Detection of Traffic Anomaly in Driving Videos , IEEE Transactions on Pattern Analysis & Machine Intelligence 45 (01) (2023) 444–459.doi:10.1109/TPAMI.2022.3150763. URL https://doi.ieeecomput...

2023
[12]

J. Yu, J. Jiang, S. Fichera, P. Paoletti, L. Layzell, D. Mehta, S. Luo, Road surface defect detection—from image-based to non-image-based: A survey, IEEE Transactions on Intelligent Transportation Systems PP (2024) 1–23.doi:10.1109/TITS.2024.3382837

2024
[13]

Chan, Y.-T

F.-H. Chan, Y.-T. Chen, Y. Xiang, M. Sun, Anticipating accidents in dashcam videos, in: Asian Conference on Computer Vision, 2016. URL https://api.semanticscholar.org/CorpusID:45520437

2016
[14]

H. Kim, K. Lee, G. Hwang, C. Suh, Crash to not crash: Learn to identify dangerous vehicles using a simulator, Proceedings of the AAAI Conference on Artificial Intelligence 33 (2019) 978–985.doi:10.1609/aaai.v33i01.3301978

2019 doi
[15]

206–213.doi:10.1109/ICCVW.2017.33

A.Rasouli,I.Kotseruba,J.K.Tsotsos,Aretheygoingtocross?abenchmarkdatasetandbaselineforpedestriancrosswalkbehavior,in:2017 IEEE International Conference on Computer Vision Workshops (ICCVW), 2017, pp. 206–213.doi:10.1109/ICCVW.2017.33

2017 doi
[16]

Pinggera, S

P. Pinggera, S. Ramos, S. Gehrig, U. Franke, C. Rother, R. Mester, Lost and found: detecting small road hazards for self-driving vehicles, in:2016IEEE/RSJInternationalConferenceonIntelligentRobotsandSystems(IROS),2016,pp.1099–1106. doi:10.1109/IROS.2016. 7759186

2016 doi
[17]

R. Chan, K. Lis, S. Uhlemeyer, H. Blum, S. Honari, R. Siegwart, P. Fua, M. Salzmann, M. Rottmann, Segmentmeifyoucan: A benchmark for anomaly segmentation, in: Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track, 2021

2021
[18]

K. Lis, K. Nakka, P. Fua, M. Salzmann, Detecting the unexpected via image resynthesis, in: 2019 IEEE/CVF International Conference on Computer Vision (ICCV), IEEE Computer Society, Los Alamitos, CA, USA, 2019, pp. 2152–2161.doi:10.1109/ICCV.2019.00224. URL https://doi.ieeecompu...

2019
[19]

2403–2412.doi:10.1109/ICCVW.2019

H.Blum,P.-E.Sarlin,J.Nieto,R.Siegwart,C.Cadena,Fishyscapes:Abenchmarkforsafesemanticsegmentationinautonomousdriving,in: 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), 2019, pp. 2403–2412.doi:10.1109/ICCVW.2019. 00294

2019 doi
[20]

A. P. Shah, J.-B. Lamare, T. Nguyen-Anh, A. Hauptmann, Cadp: A novel dataset for cctv traffic camera based accident analysis, in: 2018 15thIEEEInternationalConferenceonAdvancedVideoandSignalBasedSurveillance(AVSS),2018,pp.1–9. doi:10.1109/AVSS.2018. 8639160

2018 doi
[21]

doi:10.1145/3394171.3413827

W.Bao,Q.Yu,Y.Kong,Uncertainty-basedtrafficaccidentanticipationwithspatio-temporalrelationallearning,in:Proceedingsofthe28th ACMInternationalConferenceonMultimedia,MM’20,AssociationforComputingMachinery,NewYork,NY,USA,2020,p.2682–2690. doi:10.1145/3394171.3413827. URL https://d...

2020
[22]

Y.Yao,M.Xu,Y.Wang,D.J.Crandall,E.M.Atkins,Unsupervisedtrafficaccidentdetectioninfirst-personvideos,in:IEEE/RSJInternational Conference on Intelligent Robots and Systems (IROS), 2019

2019
[23]

: Preprint submitted to Elsevier Page 75 of 85

J.Fang,J.Qiao,J.Xue,Z.Li,Vision-basedtrafficaccidentdetectionandanticipation:Asurvey,IEEETransactionsonCircuitsandSystems for Video Technology 34 (4) (2024) 1983–1999.doi:10.1109/TCSVT.2023.3307655. : Preprint submitted to Elsevier Page 75 of 85

2024
[24]

S.-Y. Yu, A. V. Malawade, D. Muthirayan, P. P. Khargonekar, M. A. A. Faruque, Scene-graph augmented data-driven risk assessment of autonomous vehicle decisions, IEEE Transactions on Intelligent Transportation Systems 23 (7) (2022) 7941–7951.doi:10.1109/TITS. 2021.3074854

2022
[25]

N. Bao, D. Yang, A. Carballo, Ü. Özgüner, K. Takeda, Personalized safety-focused control by minimizing subjective risk, in: 2019 IEEE Intelligent Transportation Systems Conference (ITSC), IEEE, 2019, pp. 3853–3858

2019
[26]

Kopuklu, J

O. Kopuklu, J. Zheng, H. Xu, G. Rigoll, Driver anomaly detection: A dataset and contrastive learning approach, in: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2021, pp. 91–100

2021
[27]

Zhang, S

Y. Zhang, S. Wang, Z. Shi, W. Yang, A survey on anomaly segmentation in urban scene understanding with image data, Knowledge-Based Systems 338 (2026) 115521.doi:https://doi.org/10.1016/j.knosys.2026.115521. URL https://www.sciencedirect.com/science/article/pii/S0950705126002637

2026
[28]

R.Jiao,Y.Wan,F.Poiesi,Y.Wang,Surveyonvideoanomalydetectionindynamicsceneswithmovingcameras,Artif.Intell.Rev.56(Suppl
[29]

URL https://doi.org/10.1007/s10462-023-10609-x

(2023) 3515–3570.doi:10.1007/s10462-023-10609-x. URL https://doi.org/10.1007/s10462-023-10609-x

2023 doi
[30]

L. Zhu, L. Wang, A. Raj, T. Gedeon, C. Chen, Advancing video anomaly detection: a concise review and a new dataset, in: Proceedings of the38thInternationalConferenceonNeuralInformationProcessingSystems,NIPS’24,CurranAssociatesInc.,RedHook,NY,USA,2024

2024
[31]

doi:10.1145/1541880.1541882

V.Chandola,A.Banerjee,V.Kumar,Anomalydetection:Asurvey,ACMComput.Surv.41(072009). doi:10.1145/1541880.1541882
[32]

arXiv:https://doi.org/10.1080/14680629.2023.2237601, doi:10.1080/14680629.2023.2237601

C.Abdollahi,M.Mollajafari,A.Golroo,S.Moridpour,H.Wang,Areviewonpavementdataacquisitionandanalyticstoolsusingautonomous vehicles,RoadMaterialsandPavementDesign25(5)(2024)914–940. arXiv:https://doi.org/10.1080/14680629.2023.2237601, doi:10.1080/14680629.2023.2237601. URL https:/...

2024
[33]

Ghari, A

B. Ghari, A. Tourani, A. Shahbahrami, G. Gaydadjiev, Pedestrian detection in low-light conditions: A comprehensive survey, Image Vision Comput. 148 (C) (Aug. 2024).doi:10.1016/j.imavis.2024.105106. URL https://doi.org/10.1016/j.imavis.2024.105106

2024
[34]

J. Cao, Y. Pang, J. Xie, F. S. Khan, L. Shao, From handcrafted to deep features for pedestrian detection: A survey, IEEE Transactions on Pattern Analysis and Machine Intelligence 44 (9) (2022) 4913–4934.doi:10.1109/TPAMI.2021.3076733

2022
[35]

J. Chen, Q. Wang, H. H. Cheng, W. Peng, W. Xu, A review of vision-based traffic semantic understanding in itss, IEEE Transactions on Intelligent Transportation Systems 23 (11) (2022) 19954–19979.doi:10.1109/TITS.2022.3182410

2022
[36]

M.Liu,E.Yurtsever,J.Fossaert,X.Zhou,W.Zimmer,Y.Cui,B.L.Zagar,A.C.Knoll,Asurveyonautonomousdrivingdatasets:Statistics, annotation quality, and a future outlook, IEEE Transactions on Intelligent Vehicles (2024) 1–29doi:10.1109/TIV.2024.3394735

2024
[37]

Paneru, I

S. Paneru, I. Jeelani, Computer vision applications in construction: Current state, opportunities & challenges, Automation in Construction 132 (2021) 103940.doi:https://doi.org/10.1016/j.autcon.2021.103940. URL https://www.sciencedirect.com/science/article/pii/S0926580521003915

2021
[38]

Simonyan, A

K. Simonyan, A. Zisserman, Very deep convolutional networks for large-scale image recognition, in: International Conference on Learning Representations, 2015

2015
[39]

K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, in: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 770–778.doi:10.1109/CVPR.2016.90

2016 doi
[40]

Dosovitskiy, L

A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly, J. Uszkoreit, N. Houlsby, An image is worth 16x16 words: Transformers for image recognition at scale, in: International Conference on Learning Repr...

2021
[41]

doi:10.48550/arXiv.2602.23120

A.Sabaghi,J.OramasM,Trilite:Efficientweaklysupervisedobjectlocalizationwithuniversalvisualfeaturesandtri-regiondisentanglement (02 2026). doi:10.48550/arXiv.2602.23120

2026 doi
[42]

J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, L. Fei-Fei, Imagenet: A large-scale hierarchical image database, in: 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009, pp. 248–255.doi:10.1109/CVPR.2009.5206848

2009
[43]

F.Yu,H.Chen,X.Wang,W.Xian,Y.Chen,F.Liu,V.Madhavan,T.Darrell,Bdd100k:Adiversedrivingdatasetforheterogeneousmultitask learning, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020

2020
[44]

Everingham, L

M. Everingham, L. van Gool, C. Williams, J. Winn, A. Zisserman, The pascal visual object classes (voc) challenge, International Journal of Computer Vision 88 (2) (2010) 303–338.doi:10.1007/s11263-009-0275-4

2010 doi
[45]

URL https://api.semanticscholar.org/CorpusID:347907

P.Henderson,V.Ferrari,End-to-endtrainingofobjectclassdetectorsformeanaverageprecision,in:AsianConferenceonComputerVision, 2016. URL https://api.semanticscholar.org/CorpusID:347907

2016
[46]

A. D. Rhodes, M. H. Quinn, M. Mitchell, Fast on-line kernel density estimation for active object localization, in: 2017 International Joint Conference on Neural Networks (IJCNN), 2017, pp. 454–462.doi:10.1109/IJCNN.2017.7965889

2017
[47]

U. A. Computer Vision Lab, Lecture 12: Computer vision, Course Materials for COMPSCI 682, available at:https://cvl-umass. github.io/compsci682-fall-2024/docs/lecture12fall2024.pdf (2024)

2024
[48]

Conference on Computer Vision and Pattern Recognition (CVPR)

A. Nekrasov, M. Burdorf, S. Worrall, B. Leibe, J. S. B. Perez, Spotting the Unexpected (STU): A 3D LiDAR Dataset for Anomaly Segmentation in Autonomous Driving, in: "Conference on Computer Vision and Pattern Recognition (CVPR)", 2025

2025
[49]

O. O. Xiying Li, Chinese city traffic image database(cctrib) (2022). URL http://www.openits.cn/openData4/824.jhtml

2022
[50]

L. Wen, D. Du, Z. Cai, Z. Lei, M.-C. Chang, H. Qi, J. Lim, M.-H. Yang, S. Lyu, Ua-detrac: A new benchmark and protocol for multi-object detectionandtracking,ComputerVisionandImageUnderstanding193(2020)102907. doi:https://doi.org/10.1016/j.cviu.2020. 102907. URL https://www.sci...

2020 doi
[51]

URL https://dx.doi.org/10.21227/tjtg-nz28

V.Adewopo,N.Elsayed,Z.ElSayed,M.Ozer,C.Zekios,A.Abdelgawad,M.Bayoumi,Trafficaccidentdetectionvideodatasetforai-driven computer vision systems in smart city transportation (2023).doi:10.21227/tjtg-nz28. URL https://dx.doi.org/10.21227/tjtg-nz28

2023 doi
[52]

T. You, B. Han, Traffic Accident Benchmark for Causality Recognition, in: ECCV, 2020

2020
[53]

Fang, L.-l

J. Fang, L.-l. Li, J. Zhou, J. Xiao, H. Yu, C. Lv, J. Xue, T.-S. Chua, Abductive ego-view accident video understanding for safe driving perception, in: 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024, pp. 22030–22040. doi: 10.1109/CVPR52733.2024.02080

2024
[54]

Sultani, C

W. Sultani, C. Chen, M. Shah, Real-world anomaly detection in surveillance videos, in: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018

2018
[55]

Chan, Y.-T

F.-H. Chan, Y.-T. Chen, Y. Xiang, M. Sun, Anticipating accidents in dashcam videos, in: S.-H. Lai, V. Lepetit, K. Nishino, Y. Sato (Eds.), Computer Vision – ACCV 2016, Springer International Publishing, Cham, 2017, pp. 136–153

2016
[56]

Pradana, M.-S

H. Pradana, M.-S. Dao, K. Zettsu, Augmenting ego-vehicle for traffic near-miss and accident classification dataset using manipulating conditional style translation, in: 2022 International Conference on Digital Image Computing: Techniques and Applications (DICTA), 2022, pp. 1–8...

2022
[57]

Q. Zou, Y. Cao, Q. Li, Q. Mao, S. Wang, Cracktree: Automatic crack detection from pavement images, Pattern Recognition Letters 33 (3) (2012) 227–238. doi:https://doi.org/10.1016/j.patrec.2011.11.004. URL https://www.sciencedirect.com/science/article/pii/S0167865511003795

2012 doi
[58]

Y. Shi, L. Cui, Z. Qi, F. Meng, Z. Chen, Automatic road crack detection using random structured forests, IEEE Transactions on Intelligent Transportation Systems 17 (12) (2016) 3434–3445.doi:10.1109/TITS.2016.2552248

2016
[59]

F. Yang, L. Zhang, S. Yu, D. Prokhorov, X. Mei, H. Ling, Feature pyramid and hierarchical boosting network for pavement crack detection, IEEE Transactions on Intelligent Transportation Systems 21 (4) (2020) 1525–1535.doi:10.1109/TITS.2019.2910595

2020
[60]

URL https://www.sciencedirect.com/science/article/pii/S0950061820314021

Q.Mei,M.Gül,Acosteffectivesolutionforpavementcrackinspectionusingcamerasanddeepneuralnetworks,ConstructionandBuilding Materials 256 (2020) 119397.doi:https://doi.org/10.1016/j.conbuildmat.2020.119397. URL https://www.sciencedirect.com/science/article/pii/S0950061820314021

2020
[61]

Z.Huang,W.Chen,A.Al-Tabbaa,I.Brilakis,Nha12d:Anewpavementcrackdatasetandacomparisonstudyofcrackdetectionalgorithms, 2022 European Conference on Computing in Construction (2022)

2022
[62]

R. Fan, U. Ozgunalp, B. Hosking, M. Liu, I. Pitas, Pothole detection based on disparity transformation and road surface modeling, IEEE Transactions on Image Processing 29 (2020) 897–908.doi:10.1109/TIP.2019.2933750

2020
[63]

W. Tang, Q. Zhao, S. Huang, R. Li, L. Huangfu, An iteratively optimized patch label inference network for automatic pavement distress detection, IEEE Transactions on Intelligent Transportation Systems 23 (2020) 8652–8661. URL https://api.semanticscholar.org/CorpusID:218900648

2020
[64]

Moscoso Thompson, A

E. Moscoso Thompson, A. Ranieri, S. Biasotti, M. Chicchon, I. Sipiran, M.-K. Pham, T.-L. Nguyen-Ho, H.-D. Nguyen, M.-T. Tran, Shrec 2022: Pothole and crack detection in the road pavement using images and rgb-d data, Comput. Graph. 107 (C) (2022) 161–171. doi:10.1016/j.cag.2022...

2022 doi
[65]

T.-Y.Tseng,H.Lyu,J.Li,J.S.Berrio,M.Shan,S.Worrall,M2s-road:Multi-modalsemanticsegmentationforroaddamageusingcameraand lidar data, in: Proceedings of the 2024 Australasian Conference on Robotics and Automation (ACRA), Australian Robotics and Automation Association, 2024, in press

2024
[66]

Stricker, M

R. Stricker, M. Eisenbach, M. Sesselmann, K. Debes, H.-M. Gross, Improving visual road condition assessment by extensive experiments on the extended gaps dataset, in: 2019 International Joint Conference on Neural Networks (IJCNN), 2019, pp. 1–8.doi:10.1109/IJCNN. 2019.8852257

2019
[67]

arXiv:https://rmets.onlinelibrary.wiley.com/doi/pdf/10.1002/gdj3.260, doi:https://doi.org/10.1002/gdj3.260

D.Arya,H.Maeda,S.K.Ghosh,D.Toshniwal,Y.Sekimoto,Rdd2022:Amulti-nationalimagedatasetforautomaticroaddamagedetection, GeoscienceDataJournal11(4)(2024)846–862. arXiv:https://rmets.onlinelibrary.wiley.com/doi/pdf/10.1002/gdj3.260, doi:https://doi.org/10.1002/gdj3.260. URL https://...

2024 doi
[68]

Sakaridis, D

C. Sakaridis, D. Dai, L. Van Gool, ACDC: The adverse conditions dataset with correspondences for semantic driving scene understanding, in: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021

2021
[69]

X.Hu,C.-W.Fu,L.Zhu,P.-A.Heng,Depth-attentionalfeaturesforsingle-imagerainremoval,in:ProceedingsoftheIEEE/CVFConference on Computer Vision and Pattern Recognition (CVPR), 2019

2019
[70]

Sakaridis, D

C. Sakaridis, D. Dai, S. Hecker, L. Van Gool, Model adaptation with synthetic and real data for semantic dense foggy scene understanding, in: Proceedings of the European Conference on Computer Vision (ECCV), 2018

2018
[71]

URL https://doi.org/10.1007/s11263-018-1072-8

C.Sakaridis,D.Dai,L.VanGool,Semanticfoggysceneunderstandingwithsyntheticdata,InternationalJournalofComputerVision126(9) (2018) 973–992. URL https://doi.org/10.1007/s11263-018-1072-8

2018 doi
[72]

3819–3824.doi:10.1109/ITSC.2018.8569387

D.Dai,L.V.Gool,Darkmodeladaptation:Semanticimagesegmentationfromdaytimetonighttime,in:201821stInternationalConference on Intelligent Transportation Systems (ITSC), 2018, pp. 3819–3824.doi:10.1109/ITSC.2018.8569387

2018
[73]

X. Tan, K. Xu, Y. Cao, Y. Zhang, L. Ma, R. W. H. Lau, Night-time scene parsing with a large real dataset, IEEE Transactions on Image Processing 30 (2021) 9085–9098.doi:10.1109/TIP.2021.3122004

2021
[74]

C.Sakaridis,D.Dai,L.VanGool,Map-guidedcurriculumdomainadaptationanduncertainty-awareevaluationforsemanticnighttimeimage segmentation, IEEE Transactions on Pattern Analysis and Machine Intelligence 44 (6) (2022) 3139–3153.doi:10.1109/TPAMI.2020. 3045882

2022 doi
[75]

Neumann, M

L. Neumann, M. Karg, S. Zhang, C. Scharfenberger, E. Piegert, S. Mistr, O. Prokofyeva, R. Thiel, A. Vedaldi, A. Zisserman, B. Schiele, Nightowls: A pedestrians at night dataset, in: Asian Conference on Computer Vision, 2018. : Preprint submitted to Elsevier Page 77 of 85 URL h...

2018
[76]

N. Gray, M. Moraes, J. Bian, A. Wang, A. Tian, K. Wilson, Y. Huang, H. Xiong, Z. Guo, Glare: A dataset for traffic sign detection in sun glare, IEEE Transactions on Intelligent Transportation Systems 24 (11) (2023) 12323–12330.doi:10.1109/TITS.2023.3294411

2023
[78]

Rasouli, I

A. Rasouli, I. Kotseruba, T. Kunic, J. K. Tsotsos, Pie: A large-scale dataset and models for pedestrian intention estimation and trajectory prediction, in: International Conference on Computer Vision (ICCV), 2019

2019
[79]

B.Liu,E.Adeli,Z.Cao,K.-H.Lee,A.Shenoi,A.Gaidon,J.C.Niebles,Spatiotemporalrelationshipreasoningforpedestrianintentprediction, in: IEEE Robotics and Automation Letters (IEEE RA-L) and International Conference on Robotics and Automation (ICRA), IEEE, 2020

2020
[80]

W.Kim,M.S.Ramanagopal,C.Barto,M.-Y.Yu,K.Rosaen,N.Goumas,R.Vasudevan,M.Johnson-Roberson,Pedx:Benchmarkdatasetfor metric 3-d pose estimation of pedestrians in complex urban intersections, IEEE Robotics and Automation Letters 4 (2) (2019) 1940–1947

2019
[82]

Braun, F

M. Braun, F. B. Flohr, S. Krebs, U. Kreße, D. M. Gavrila, Simple pair pose - pairwise human pose estimation in dense urban traffic scenes, in: 2021 IEEE Intelligent Vehicles Symposium (IV), 2021, pp. 1545–1552.doi:10.1109/IV48863.2021.9575435

2021
[83]

A. V. Malawade, S.-Y. Yu, B. Hsu, D. Muthirayan, P. P. Khargonekar, M. A. A. Faruque, Spatiotemporal scene-graph embedding for autonomousvehiclecollisionprediction,IEEEInternetofThingsJournal9(12)(2022)9379–9388. doi:10.1109/JIOT.2022.3141044

2022
[84]

Yurtsever, Y

E. Yurtsever, Y. Liu, J. Lambert, C. Miyajima, E. Takeuchi, K. Takeda, J. H. L. Hansen, Risky action recognition in lane change video clips usingdeepspatiotemporalnetworkswithsegmentationmasktransfer,in:2019IEEEIntelligentTransportationSystemsConference(ITSC), 2019, pp. 3100–3...

2019
[85]

Dollar, C

P. Dollar, C. Wojek, B. Schiele, P. Perona, Pedestrian detection: A benchmark, in: 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009, pp. 304–311.doi:10.1109/CVPR.2009.5206631

2009
[86]

Zhang, R

S. Zhang, R. Benenson, B. Schiele, Citypersons: A diverse dataset for pedestrian detection, in: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 4457–4465.doi:10.1109/CVPR.2017.474

2017 doi
[87]

M.Braun,S.Krebs,F.B.Flohr,D.M.Gavrila,Eurocitypersons:Anovelbenchmarkforpersondetectionintrafficscenes,IEEETransactions on Pattern Analysis and Machine Intelligence (2019) 1–1doi:10.1109/TPAMI.2019.2897684

2019
[88]

X. Jia, C. Zhu, M. Li, W. Tang, W. Zhou, LLVIP: A Visible-infrared Paired Dataset for Low-light Vision , in: 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), IEEE Computer Society, Los Alamitos, CA, USA, 2021, pp. 3489–3497.doi: 10.1109/ICCVW54120.2...

2021
[89]

X. Li, F. Flohr, Y. Yang, H. Xiong, M. Braun, S. Pan, K. Li, D. M. Gavrila, A new benchmark for vision-based cyclist detection, in: 2016 IEEE Intelligent Vehicles Symposium (IV), 2016, pp. 1028–1033.doi:10.1109/IVS.2016.7535515

2016
[90]

K. Zhou, C. Li, M. Wang, L. Tomato, Tusimple competitions for cvpr2017,https://github.com/TuSimple/tusimple-benchmark (2017)

2017
[91]

X. Pan, J. Shi, P. Luo, X. Wang, X. Tang, Spatial as deep: Spatial cnn for traffic scene understanding, in: AAAI Conference on Artificial Intelligence (AAAI), 2018

2018
[92]

Zhang, L

Y. Zhang, L. Zhu, W. Feng, H. Fu, M. Wang, Q. Li, C. Li, S. Wang, Vil-100: A new dataset and a baseline model for video instance lane detection, 2021 IEEE/CVF International Conference on Computer Vision (ICCV) (2021) 15661–15670. URL https://api.semanticscholar.org/CorpusID:237213300

2021
[93]

L. Chen, C. Sima, Y. Li, Z. Zheng, J. Xu, X. Geng, H. Li, C. He, J. Shi, Y. Qiao, J. Yan, Persformer: 3d lane detection via perspective transformer and the openlane benchmark, in: European Conference on Computer Vision (ECCV), 2022

2022
[94]

Huang, P

X. Huang, P. Wang, X. Cheng, D. Zhou, Q. Geng, R. Yang, The apolloscape open dataset for autonomous driving and its application, IEEE Transactions on Pattern Analysis and Machine Intelligence 42 (10) (2020) 2702–2719.doi:10.1109/TPAMI.2019.2926463

2020
[97]

B.Wilson,W.Qi,T.Agarwal,J.Lambert,J.Singh,S.Khandelwal,B.Pan,R.Kumar,A.Hartnett,J.K.Pontes,D.Ramanan,P.Carr,J.Hays, Argoverse 2: Next generation datasets for self-driving perception and forecasting, in: Proceedings of the Neural Information Processing Systems Track on Datasets...

2021
[98]

D.Tabernik,D.Skočaj,Deeplearningforlarge-scaletraffic-signdetectionandrecognition,IEEETransactionsonIntelligentTransportation Systems 21 (4) (2020) 1427–1440.doi:10.1109/TITS.2019.2913588

2020
[99]

Z.Zhu,D.Liang,S.Zhang,X.Huang,B.Li,S.Hu,Traffic-signdetectionandclassificationinthewild,in:TheIEEEConferenceonComputer Vision and Pattern Recognition (CVPR), 2016

2016
[100]

Gámez Serna, Y

C. Gámez Serna, Y. Ruichek, Traffic signs detection and classification for european urban environments, IEEE Transactions on Intelligent Transportation Systems 21 (10) (2020) 4388–4399.doi:10.1109/TITS.2019.2941081

2020
[101]

Y. Guo, W. Feng, F. Yin, T. Xue, S. Mei, C.-L. Liu, Learning to understand traffic signs, in: Proceedings of the 29th ACM International Conference on Multimedia, MM ’21, Association for Computing Machinery, New York, NY, USA, 2021, p. 2076–2084.doi:10.1145/ : Preprint submitte...

2021
[103]

M. B. Jensen, M. P. Philipsen, A. Møgelmose, T. B. Moeslund, M. M. Trivedi, Vision for looking at traffic lights: Issues, survey, and perspectives, IEEE Transactions on Intelligent Transportation Systems 17 (7) (2016) 1800–1815.doi:10.1109/TITS.2015.2509509

2016
[104]

Behrendt, L

K. Behrendt, L. Novak, R. Botros, A deep learning approach to traffic lights: Detection, tracking, and classification, in: 2017 IEEE International Conference on Robotics and Automation (ICRA), 2017, pp. 1370–1377.doi:10.1109/ICRA.2017.7989163

2017
[105]

Fregin, J

A. Fregin, J. Muller, U. Krebel, K. Dietmayer, The driveu traffic light dataset: Introduction and comparison with existing datasets, in: 2018 IEEE International Conference on Robotics and Automation (ICRA), 2018, pp. 3376–3383.doi:10.1109/ICRA.2018.8460737

2018
[106]

X. Yang, J. Yan, W. Liao, X. Yang, J. Tang, T. He, SCRDet++: Detecting Small, Cluttered and Rotated Objects via Instance-Level Feature Denoising and Rotation Loss Smoothing , IEEE Transactions on Pattern Analysis & Machine Intelligence 45 (02) (2023) 2384–2399. doi:10.1109/TPA...

2023
[107]

J. He, C. Zhang, X. He, R. Dong, Visual recognition of traffic police gestures with convolutional pose machine and handcrafted features, Neurocomputing 390 (2020) 248–259.doi:https://doi.org/10.1016/j.neucom.2019.07.103. URL https://www.sciencedirect.com/science/article/pii/S0...

2020 doi
[108]

Izquierdo, A

R. Izquierdo, A. Quintanar, I. Parra, D. Fernández-Llorca, M. A. Sotelo, The prevention dataset: a novel benchmark for prediction of vehicles intentions, in: 2019 IEEE Intelligent Transportation Systems Conference (ITSC), 2019, pp. 3114–3121.doi:10.1109/ITSC. 2019.8917433

2019
[109]

International Conference on Robotics and Automation, in press, 2019

J.Xue,J.Fang,T.Li,B.Zhang,P.Zhang,Z.Ye,J.Dou,BLVD:Buildingalarge-scale5dsemanticsbenchmarkforautonomousdriving,in: Proc. International Conference on Robotics and Automation, in press, 2019

2019
[110]

Ettinger, S

S. Ettinger, S. Cheng, B. Caine, C. Liu, H. Zhao, S. Pradhan, Y. Chai, B. Sapp, C. R. Qi, Y. Zhou, Z. Yang, A. Chouard, P. Sun, J. Ngiam, V. Vasudevan, A. McCauley, J. Shlens, D. Anguelov, Large scale interactive motion forecasting for autonomous driving: The waymo open motion...

2021
[111]

J. Choe, S. J. Oh, S. Chun, S. Lee, Z. Akata, H. Shim, Evaluation for weakly supervised object localization: Protocol, metrics, and datasets, IEEE Transactions on Pattern Analysis and Machine Intelligence 45 (2) (2023) 1732–1748.doi:10.1109/TPAMI.2022.3169881

2023
[112]

Gupta, S

S. Gupta, S. Lakhotia, A. Rawat, R. Tallamraju, Vitol: Vision transformer for weakly supervised object localization, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 4101–4110

2022
[113]

O.Russakovsky,J.Deng,H.Su,J.Krause,S.Satheesh,S.Ma,Z.Huang,A.Karpathy,A.Khosla,M.Bernstein,A.Berg,L.Fei-Fei,Imagenet large scale visual recognition challenge, International Journal of Computer Vision 115 (09 2014).doi:10.1007/s11263-015-0816-y

2014 doi
[114]

K. He, G. Gkioxari, P. Dollár, R. B. Girshick, Mask r-cnn, 2017 IEEE International Conference on Computer Vision (ICCV) (2017) 2980– 2988. URL https://api.semanticscholar.org/CorpusID:206771194

2017
[115]

S.Ren,K.He,R.Girshick,J.Sun,Fasterr-cnn:Towardsreal-timeobjectdetectionwithregionproposalnetworks,in:C.Cortes,N.Lawrence, D. Lee, M. Sugiyama, R. Garnett (Eds.), Advances in Neural Information Processing Systems, Vol. 28, Curran Associates, Inc., 2015. URL https://proceedings....

2015
[116]

Redmon, S

J. Redmon, S. K. Divvala, R. B. Girshick, A. Farhadi, You only look once: Unified, real-time object detection, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016) 779–788. URL https://api.semanticscholar.org/CorpusID:206594738

2016
[117]

Welling (Eds.), Computer Vision – ECCV 2016, Springer International Publishing, Cham, 2016, pp

W.Liu,D.Anguelov,D.Erhan,C.Szegedy,S.Reed,C.-Y.Fu,A.C.Berg,Ssd:Singleshotmultiboxdetector,in:B.Leibe,J.Matas,N.Sebe, M. Welling (Eds.), Computer Vision – ECCV 2016, Springer International Publishing, Cham, 2016, pp. 21–37

2016
[118]

Z. Tian, C. Shen, H. Chen, T. He, Fcos: Fully convolutional one-stage object detection, in: 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019, pp. 9626–9635.doi:10.1109/ICCV.2019.00972

2019
[119]

N.Carion,F.Massa,G.Synnaeve,N.Usunier,A.Kirillov,S.Zagoruyko,End-to-endobjectdetectionwithtransformers,in:ComputerVision –ECCV2020:16thEuropeanConference,Glasgow,UK,August23–28,2020,Proceedings,PartI,Springer-Verlag,Berlin,Heidelberg,2020, p. 213–229. doi:10.1007/978-3-030-584...

2020 doi
[120]

Robinson, P

I. Robinson, P. Robicheaux, M. Popov, D. Ramanan, N. Peri, Rf-detr: Neural architecture search for real-time detection transformers (2025). arXiv:2511.09554. URL https://arxiv.org/abs/2511.09554

2025
[121]

Everingham, S

M. Everingham, S. M. A. Eslami, L. Van Gool, C. K. I. Williams, J. Winn, A. Zisserman, The pascal visual object classes challenge: A retrospective, International Journal of Computer Vision 111 (1) (2015) 98–136

2015
[122]

URL https://api.semanticscholar.org/CorpusID:14113767

T.-Y.Lin,M.Maire,S.J.Belongie,J.Hays,P.Perona,D.Ramanan,P.Dollár,C.L.Zitnick,Microsoftcoco:Commonobjectsincontext,in: European Conference on Computer Vision, 2014. URL https://api.semanticscholar.org/CorpusID:14113767

2014
[123]

of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016

M.Cordts,M.Omran,S.Ramos,T.Rehfeld,M.Enzweiler,R.Benenson,U.Franke,S.Roth,B.Schiele,Thecityscapesdatasetforsemantic urban scene understanding, in: Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016

2016
[124]

Girshick, Fast r-cnn, in: 2015 IEEE International Conference on Computer Vision (ICCV), 2015, pp

R. Girshick, Fast r-cnn, in: 2015 IEEE International Conference on Computer Vision (ICCV), 2015, pp. 1440–1448.doi:10.1109/ICCV. 2015.169

2015 doi
[126]

Ronneberger, P

O. Ronneberger, P. Fischer, T. Brox, U-net: Convolutional networks for biomedical image segmentation, in: N. Navab, J. Hornegger, W. M. Wells, A. F. Frangi (Eds.), Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, Springer International Publishing, Cham...

2015
[127]

L.-C. Chen, G. Papandreou, I. Kokkinos, K. Murphy, A. L. Yuille, Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs, IEEE Transactions on Pattern Analysis and Machine Intelligence 40 (4) (2018) 834–848. doi:10.1109/T...

2018
[128]

Zheng, J

S. Zheng, J. Lu, H. Zhao, X. Zhu, Z. Luo, Y. Wang, Y. Fu, J. Feng, T. Xiang, P. H. Torr, L. Zhang, Rethinking semantic segmentation from a sequence-to-sequence perspective with transformers, in: CVPR, 2021

2021
[129]

E. Xie, W. Wang, Z. Yu, A. Anandkumar, J. M. Alvarez, P. Luo, Segformer: Simple and efficient design for semantic segmentation with transformers, in: Neural Information Processing Systems (NeurIPS), 2021

2021
[130]

5000–5009.doi:10.1109/ICCV.2017.534

G.Neuhold,T.Ollmann,S.R.Bulò,P.Kontschieder,Themapillaryvistasdatasetforsemanticunderstandingofstreetscenes,in:2017IEEE International Conference on Computer Vision (ICCV), 2017, pp. 5000–5009.doi:10.1109/ICCV.2017.534

2017 doi
[131]

Hariharan, P

B. Hariharan, P. Arbeláez, R. Girshick, J. Malik, Simultaneous detection and segmentation, in: European Conference on Computer Vision (ECCV), 2014

2014
[132]

X. Wang, T. Kong, C. Shen, Y. Jiang, L. Li, Solo: Segmenting objects by locations, in: Proc. Eur. Conf. Computer Vision (ECCV), 2020

2020
[133]

Kirillov, Y

A. Kirillov, Y. Wu, K. He, R. B. Girshick, Pointrend: Image segmentation as rendering, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2019) 9796–9805. URL https://api.semanticscholar.org/CorpusID:209386851

2019
[134]

B. Dong, F. Zeng, T. Wang, X. Zhang, Y. Wei, Solq: segmenting objects by learning queries, in: Proceedings of the 35th International Conference on Neural Information Processing Systems, NIPS ’21, Curran Associates Inc., Red Hook, NY, USA, 2021

2021
[135]

Cheng, I

B. Cheng, I. Misra, A. G. Schwing, A. Kirillov, R. Girdhar, Masked-attention mask transformer for universal image segmentation, in: 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 1280–1289.doi:10.1109/CVPR52688.2022. 00135

2022
[138]

Kirillov, R

A. Kirillov, R. Girshick, K. He, P. Dollár, Panoptic feature pyramid networks, in: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 6392–6401.doi:10.1109/CVPR.2019.00656

2019
[139]

12472–12482

B.Cheng,M.D.Collins,Y.Zhu,T.Liu,T.S.Huang,H.Adam,L.-C.Chen,Panoptic-deeplab:Asimple,strong,andfastbaselineforbottom- up panoptic segmentation, in: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, pp. 12472–12482. doi:10.1109/CVPR42600.2020.01249

2020
[140]

Chang, S.-E

C.-Y. Chang, S.-E. Chang, P. Hsiao, L. Fu, Epsnet: Efficient panoptic segmentation network with cross-layer attention fusion, ArXiv abs/2003.10142 (2020)

2020 arXiv
[141]

J. Jain, J. Li, M. Chiu, A. Hassani, N. Orlov, H. Shi, OneFormer: One Transformer to Rule Universal Image Segmentation, 2023

2023
[142]

T.-J. Yang, M. D. Collins, Y. Zhu, J.-J. Hwang, T. Liu, X. Zhang, V. Sze, G. Papandreou, L.-C. Chen, Deeperlab: Single-shot image parser, ArXiv abs/1902.05093 (2019). URL https://api.semanticscholar.org/CorpusID:61153476

2019 arXiv
[143]

D. Xu, Y. Zhu, C. Choy, L. Fei-Fei, Scene graph generation by iterative message passing, in: Computer Vision and Pattern Recognition (CVPR), 2017

2017
[144]

Johnson, R

J. Johnson, R. Krishna, M. Stark, L.-J. Li, D. A. Shamma, M. S. Bernstein, L. Fei-Fei, Image retrieval using scene graphs, in: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015, pp. 3668–3678.doi:10.1109/CVPR.2015.7298990

2015
[145]

Zareian, S

A. Zareian, S. Karaman, S.-F. Chang, Bridging knowledge graphs to generate scene graphs, in: Proceedings of the European conference on computer vision (ECCV), 2020

2020
[146]

R. Li, S. Zhang, B. Wan, X. He, Bipartite graph network with adaptive message passing for unbiased scene graph generation, in: 2021 IEEE/CVFConferenceonComputerVisionandPatternRecognition(CVPR),2021,pp.11104–11114. doi:10.1109/CVPR46437.2021. 01096

2021
[147]

Y. Cong, M. Y. Yang, B. Rosenhahn, Reltr: Relation transformer for scene graph generation, IEEE Transactions on Pattern Analysis and Machine Intelligence (2023)

2023
[148]

24229–24238

J.Im,J.Nam,N.Park,H.Lee,S.Park,Egtr:Extractinggraphfromtransformerforscenegraphgeneration,in:ProceedingsoftheIEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024, pp. 24229–24238

2024
[149]

Zhang, S

C. Zhang, S. Stepputtis, J. Campbell, K. Sycara, Y. Xie, Hiker-sgg: Hierarchical knowledge enhanced robust scene graph generation, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 28233–28243

2024
[150]

L. Li, C. ZHANG, D. Zhang, C. Sun, C. Li, L. Chen, Interaction-centric knowledge infusion and transfer for open vocabulary scene graph generation, in: The Thirty-ninth Annual Conference on Neural Information Processing Systems, 2025. URL https://openreview.net/forum?id=h2kwURAFkJ

2025
[151]

Krishna, Y

R. Krishna, Y. Zhu, O. Groth, J. Johnson, K. Hata, J. Kravitz, S. Chen, Y. Kalantidis, L.-J. Li, D. A. Shamma, M. S. Bernstein, L. Fei-Fei, Visualgenome:Connectinglanguageandvisionusingcrowdsourceddenseimageannotations,Int.J.Comput.Vision123(1)(2017)32–73. doi:10.1007/s11263-0...

2017 doi
[152]

Ferrari, The open images dataset v4, International Journal of Computer Vision 128 (2018) 1956 – 1981

A.Kuznetsova,H.Rom,N.G.Alldrin,J.R.R.Uijlings,I.Krasin,J.Pont-Tuset,S.Kamali,S.Popov,M.Malloci,A.Kolesnikov,T.Duerig, V. Ferrari, The open images dataset v4, International Journal of Computer Vision 128 (2018) 1956 – 1981. URL https://api.semanticscholar.org/CorpusID:53296866

2018
[153]

D. A. Hudson, C. D. Manning, GQA: A New Dataset for Real-World Visual Reasoning and Compositional Question Answering , in: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE Computer Society, Los Alamitos, CA, USA, 2019, pp. 6693–6702. doi:10.1109...

2019
[154]

C.Lu,R.Krishna,M.Bernstein,L.Fei-Fei,Visualrelationshipdetectionwithlanguagepriors,in:EuropeanConferenceonComputerVision, 2016

2016
[155]

T. Chen, W. Yu, R. Chen, L. Lin, Knowledge-embedded routing network for scene graph generation, in: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 6156–6164.doi:10.1109/CVPR.2019.00632

2019
[156]

T. Yin, X. Zhou, P. Krähenbühl, Center-based 3d object detection and tracking, CVPR (2021)

2021
[157]

R. Qian, X. Lai, X. Li, 3d object detection for autonomous driving: A survey, Pattern Recognition 130 (2022) 108796.doi:https: //doi.org/10.1016/j.patcog.2022.108796. URL https://www.sciencedirect.com/science/article/pii/S0031320322002771

2022
[158]

C. Nie, Z. Ju, Z. Sun, H. Zhang, 3d object detection and tracking based on lidar-camera fusion and imm-ukf algorithm towards highway driving, IEEE Transactions on Emerging Topics in Computational Intelligence 7 (4) (2023) 1242–1252.doi:10.1109/TETCI.2023. 3259441

2023 doi
[159]

6520–6530

F.Pu,Y.Wang,J.Deng,W.Yang,Monodgp:Monocular3dobjectdetectionwithdecoupled-queryandgeometry-errorpriors,in:Proceedings of the Computer Vision and Pattern Recognition Conference, 2025, pp. 6520–6530

2025
[160]

Y. Liu, J. Yan, F. Jia, S. Li, Q. Gao, T. Wang, X. Zhang, J. Sun, Petrv2: A unified framework for 3d perception from multi-camera images, arXiv preprint arXiv:2206.01256 (2022)

2022 arXiv
[161]

Y.Chen,J.Liu,X.Zhang,X.Qi,J.Jia,Voxelnext:Fullysparsevoxelnetfor3dobjectdetectionandtracking,in:ProceedingsoftheIEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023

2023
[162]

Z. Liu, H. Tang, A. Amini, X. Yang, H. Mao, D. Rus, S. Han, Bevfusion: Multi-task multi-sensor fusion with unified bird’s-eye view representation, in: IEEE International Conference on Robotics and Automation (ICRA), 2023

2023
[163]

J. Yin, J. Shen, R. Chen, W. Li, R. Yang, P. Frossard, W. Wang, Is-fusion: Instance-scene collaborative fusion for multimodal 3d object detection, in: CVPR, 2024

2024
[164]

Geiger, P

A. Geiger, P. Lenz, R. Urtasun, Are we ready for autonomous driving? the kitti vision benchmark suite, in: Conference on Computer Vision and Pattern Recognition (CVPR), 2012

2012
[167]

URL https://www.sciencedirect.com/science/article/pii/S1566253524004494

H.Xu,J.Chen,S.Meng,Y.Wang,L.-P.Chau,Asurveyonoccupancyperceptionforautonomousdriving:Theinformationfusionperspective, Information Fusion 114 (2025) 102671.doi:https://doi.org/10.1016/j.inffus.2024.102671. URL https://www.sciencedirect.com/science/article/pii/S1566253524004494

2025
[168]

A.-Q. Cao, R. de Charette, Monoscene: Monocular 3d semantic scene completion, in: CVPR, 2022

2022
[170]

Wolters, J

P. Wolters, J. Gilg, T. Teepe, F. Herzog, A. Laouichi, M. Hofmann, G. Rigoll, Unleashing hydra: Hybrid fusion, depth consistency and radar for unified 3d perception, in: 2025 IEEE International Conference on Robotics and Automation (ICRA), 2025, pp. 7467–7474. doi:10.1109/ICRA...

2025
[171]

Z. Ming, J. Berrio, M. Shan, Y. Huang, H. Lyu, N. Tran, T.-Y. Tseng, S. Worrall, Occcylindrical: Multi-modal fusion with cylindrical representation for 3d semantic occupancy prediction, in: ITSC, 2025.doi:10.48550/arXiv.2505.03284

2025 doi arXiv
[172]

Z. Ming, J. Stephany Berrio, M. Shan, S. Worrall, Occfusion: Multi-sensor fusion framework for 3d semantic occupancy prediction, IEEE Transactions on Intelligent Vehicles 10 (5) (2025) 3421–3433.doi:10.1109/TIV.2024.3453293

2025
[173]

X. Tian, T. Jiang, L. Yun, Y. Mao, H. Yang, Y. Wang, Y. Wang, H. Zhao, Occ3d: a large-scale 3d occupancy prediction benchmark for autonomous driving, in: Proceedings of the 37th International Conference on Neural Information Processing Systems, NIPS ’23, Curran Associates Inc....

2023
[174]

Y. Wei, L. Zhao, W. Zheng, Z. Zhu, J. Zhou, J. Lu, Surroundocc: Multi-camera 3d occupancy prediction for autonomous driving, 2023 IEEE/CVF International Conference on Computer Vision (ICCV) (2023) 21672–21683. URL https://api.semanticscholar.org/CorpusID:257557568

2023
[175]

17804–17813.doi:10.1109/ICCV51070.2023.01636

X.Wang,Z.Zhu,W.Xu,Y.Zhang,Y.Wei,X.Chi,Y.Ye,D.Du,J.Lu,X.Wang, OpenOccupancy:ALargeScaleBenchmarkforSurrounding Semantic Occupancy Perception , in: 2023 IEEE/CVF International Conference on Computer Vision (ICCV), IEEE Computer Society, Los Alamitos, CA, USA, 2023, pp. 17804–178...

2023
[176]

Everingham, L

M. Everingham, L. V. Gool, C. K. I. Williams, J. M. Winn, A. Zisserman, The pascal visual object classes (voc) challenge, International Journal of Computer Vision 88 (2010) 303–338. URL https://api.semanticscholar.org/CorpusID:4246903 : Preprint submitted to Elsevier Page 81 of 85

2010
[177]

4303–4309.doi:10.1109/ITSC.2019.8917218

J.Fang,D.Yan,J.Qiao,J.Xue,H.Wang,S.Li,Dada-2000:Candrivingaccidentbepredictedbydriverattention ƒanalyzedbyabenchmark, in: 2019 IEEE Intelligent Transportation Systems Conference (ITSC), 2019, pp. 4303–4309.doi:10.1109/ITSC.2019.8917218

2000
[178]

2039–2047

M.Eisenbach,R.Stricker,D.Seichter,K.Amende,K.Debes,M.Sesselmann,D.Ebersbach,U.Stoeckert,H.-M.Gross,Howtogetpavement distressdetectionreadyfordeeplearning?asystematicapproach,in:2017InternationalJointConferenceonNeuralNetworks(IJCNN),2017, pp. 2039–2047. doi:10.1109/IJCNN.2017.7966101

2017
[179]

URL https://www.sciencedirect.com/science/article/pii/S0952197621002542

P.S.Perumal,M.Sujasree,S.Chavhan,D.Gupta,V.Mukthineni,S.R.Shimgekar,A.Khanna,G.Fortino,Aninsightintocrashavoidanceand overtaking advice systems for autonomous vehicles: A review, challenges and solutions, Engineering Applications of Artificial Intelligence 104 (2021) 104406.do...

2021
[180]

Bogdoll, S

D. Bogdoll, S. Uhlemeyer, K. Kowol, J. M. Zöllner, Perception Datasets for Anomaly Detection in Autonomous Driving: A Survey, in: Intelligent Vehicles Symposium (IV), 2023

2023
[181]

K. Lis, S. Honari, P. Fua, M. Salzmann, Detecting road obstacles by erasing them, IEEE Trans. Pattern Anal. Mach. Intell. 46 (4) (2023) 2450–2460. doi:10.1109/TPAMI.2023.3335152. URL https://doi.org/10.1109/TPAMI.2023.3335152

2023
[182]

Nekrasov, A

A. Nekrasov, A. Hermans, L. Kuhnert, B. Leibe, UGainS: Uncertainty Guided Anomaly Instance Segmentation, in: GCPR, 2023

2023
[183]

M. Park, D. Q. Tran, J. Bak, S. Park, Small and overlapping worker detection at construction sites, Automation in Construction 151 (2023) 104856. doi:https://doi.org/10.1016/j.autcon.2023.104856. URL https://www.sciencedirect.com/science/article/pii/S0926580523001164

2023
[184]

A.Xuehui,Z.Li,L.Zuguang,W.Chengzhi,L.Pengfei,L.Zhiwei,Datasetandbenchmarkfordetectingmovingobjectsinconstructionsites, Automation in Construction 122 (2021) 103482

2021
[185]

R. Feng, Y. Miao, J. Zheng, A yolo-based intelligent detection algorithm for risk assessment of construction sites, Journal of Intelligent Construction (2024). doi:10.26599/JIC.2024.9180037. URL https://www.sciopen.com/article/10.26599/JIC.2024.9180037

2024
[186]

R. Duan, H. Deng, M. Tian, Y. Deng, J. Lin, Soda: A large-scale open site object detection dataset for deep learning in construction, Automation in Construction 142 (2022) 104499

2022
[187]

X. Ke, L. Shi, W. Guo, D. Chen, Multi-dimensional traffic congestion detection based on fusion of visual features and convolutional neural network, IEEE Transactions on Intelligent Transportation Systems 20 (6) (2019) 2157–2170.doi:10.1109/TITS.2018.2864612

2019
[188]

S. Jiang,Y. Feng, W.Zhang, X. Liao, X.Dai, B. O.Onasanya, A new multi-branchconvolutional neural networkand feature mapextraction method for traffic congestion detection, Sensors 24 (13) (2024).doi:10.3390/s24134272. URL https://www.mdpi.com/1424-8220/24/13/4272

2024 doi
[189]

X. Li, T. Hao, X. Jin, B. Huang, J. Liang, Fine traffic congestion detection with hierarchical description, IEEE Transactions on Intelligent Transportation Systems 23 (12) (2022) 24439–24453.doi:10.1109/TITS.2022.3206583

2022
[190]

A.Shah,J.B.Lamare,N.-A.Tuan,A.Hauptmann,Accidentforecastingincctvtrafficcameravideos,arXivpreprintarXiv:1809.05782First three authors share the first authorship. (2018)

2018 arXiv
[191]

A.P.Shah,J.-B.Lamare,T.Nguyen-Anh,A.Hauptmann,Caraccidentsdetectionandpredictiondatasetv1, https://docs.google.com/ document/d/12F7l4yxNzzUAISZufEd9WFhQKSefVVo_QsPdTsWxZh8/edit?tab=t.0 (2018)

2018
[192]

W. Bao, Q. Yu, Y. Kong, DRIVE: Deep Reinforced Accident Anticipation with Visual Explanation , in: 2021 IEEE/CVF International Conference on Computer Vision (ICCV), IEEE Computer Society, Los Alamitos, CA, USA, 2021, pp. 7599–7608.doi:10.1109/ ICCV48922.2021.00752. URL https:/...

2021
[193]

Y. Yao, M. Xu, C. Choi, D. J. Crandall, E. M. Atkins, B. Dariush, Egocentric vision-based future vehicle localization for intelligent driving assistance systems, in: International Conference on Robotics and Automation, 2019

2019
[194]

G. Tang, H. Zhao, B. Yu, Low-cost and high-performance abnormal trajectory detection based on the gru model with deep spatiotemporal sequence analysis in cloud computing, J. Cloud Comput. 13 (1) (Mar. 2024).doi:10.1186/s13677-024-00611-1. URL https://doi.org/10.1186/s13677-024-00611-1

2024 doi
[195]

A. R. Alozi, M. Hussein, How do active road users act around autonomous vehicles? an inverse reinforcement learning approach, Transportation Research Part C: Emerging Technologies 161 (2024) 104572.doi:https://doi.org/10.1016/j.trc.2024.104572. URL https://www.sciencedirect.co...

2024
[196]

Palazzi, D

A. Palazzi, D. Abati, F. Solera, R. Cucchiara, Predicting the driver’s focus of attention: the dr (eye) ve project, IEEE transactions on pattern analysis and machine intelligence 41 (7) (2018) 1720–1733

2018
[197]

Y. Liu, M. Wang, P. Lasang, Q. Sun, Importance biased traffic scene segmentation in diverse weather conditions, IEEE Transactions on Intelligent Vehicles 9 (1) (2024) 2753–2765.doi:10.1109/TIV.2023.3272922

2024
[198]

Q. Zou, Z. Zhang, Q. Li, X. Qi, Q. Wang, S. Wang, Deepcrack: Learning hierarchical convolutional features for crack detection, IEEE Transactions on Image Processing 28 (3) (2019) 1498–1512

2019
[199]

Q. Mei, M. Gül, M. R. Azim, Densely connected deep neural network considering connectivity of pixels for automatic crack detection, Automation in Construction 110 (2019) 103018.doi:10.1016/j.autcon.2019.103018

2019
[200]

doi:10.1109/TITS.2019.2931297

A.Dhiman,R.Klette,Potholedetectionusingcomputervisionandlearning,IEEETransactionsonIntelligentTransportationSystems21(8) (2020) 3536–3550. doi:10.1109/TITS.2019.2931297

2020
[201]

B. T. Passos, M. J. Cassaniga, A. M. R. Fernandes, K. B. Medeiros, E. Comunello, Cracks and potholes in road images,https://data. mendeley.com/datasets/t576ydh9v8/4 (2020)

2020
[202]

doi:10.1109/BigData55660.2022

D.Arya,H.Maeda,S.K.Ghosh,D.Toshniwal,H.Omata,T.Kashiyama,Y.Sekimoto,Crowdsensing-basedroaddamagedetectionchallenge (crddc’2022),in:2022IEEEInternationalConferenceonBigData(BigData),2022,pp.6378–6386. doi:10.1109/BigData55660.2022. 10021040. : Preprint submitted to Elsevier Pag...

2022
[203]

T. Yu, Q. Kuang, J. Hu, J. Zheng, X. Li, Global-similarity local-salience network for traffic weather recognition, IEEE Access 9 (2021) 4607–4615. doi:10.1109/ACCESS.2020.3048116

2021
[204]

Zendel, K

O. Zendel, K. Honauer, M. Murschitz, D. Steininger, G. F. Dominguez, Wilddash - creating hazard-aware benchmarks, in: Proceedings of the European Conference on Computer Vision (ECCV), 2018

2018
[205]

C. Qian, Y. Guo, Y. Mo, W. Li, Weatherdg: Llm-assisted procedural weather generation for domain-generalized semantic segmentation, https://arxiv.org/abs/2410.12075 (2024)

2024 arXiv
[206]

Toney, C

G. Toney, C. Bhargava, Adaptive headlamps in automobile: A review on the models, detection techniques, and mathematical models, IEEE Access 9 (2021) 87462–87474.doi:10.1109/ACCESS.2021.3088036

2021
[207]

Sakaridis, D

C. Sakaridis, D. Dai, L. Van Gool, Guided curriculum model adaptation and uncertainty-aware evaluation for semantic nighttime image segmentation, in: 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019, pp. 7373–7382.doi:10.1109/ICCV. 2019.00747

2019
[209]

L. A. Zebrowitz, J. M. Montepare, Social psychological face perception: Why appearance matters., Social and personality psychology compass 2 3 (2008) 1497. URL https://api.semanticscholar.org/CorpusID:11974402

2008
[210]

Verma, S

H. Verma, S. Lotia, A. Singh, Convolutional neural network based criminal detection, in: 2020 IEEE REGION 10 CONFERENCE (TENCON), 2020, pp. 1124–1129.doi:10.1109/TENCON50793.2020.9293926

2020
[211]

Singh, Facial-recognition-for-crime-detection, https://github.com/Navu4/Facial-Recognition-for-Crime-Detection (2021)

N. Singh, Facial-recognition-for-crime-detection, https://github.com/Navu4/Facial-Recognition-for-Crime-Detection (2021)

2021
[212]

Anoop, H

A. Anoop, H. G, K. Nair, S. B, V. Praseedalekshmi, T. S. H, Traffic surveillance system and criminal detection using image processing and deep learning, in: 2022 International Conference on Innovations in Science and Technology for Sustainable Development (ICISTSD), 2022, pp. ...

2022
[213]

1–5.doi:10.1109/ICoNSIP49665.2022.10007501

V.Jagtap,D.Rajmane,V.N.More,R.C.Mahajan,Suspiciousvehiclerecognitionusingnumberplate,in:2022InternationalConferenceon Signal and Information Processing (IConSIP), 2022, pp. 1–5.doi:10.1109/ICoNSIP49665.2022.10007501

2022
[214]

contributors, Automatic number-plate recognition, https://en.wikipedia.org/w/index.php?title=Automatic_ number-plate_recognition&oldid=1255542574 (2024)

W. contributors, Automatic number-plate recognition, https://en.wikipedia.org/w/index.php?title=Automatic_ number-plate_recognition&oldid=1255542574 (2024)

2024
[215]

Ayman, Automated-car-damage-detection,https://github.com/basel-ay/Automated-Car-Damage-Detection (2023)

B. Ayman, Automated-car-damage-detection,https://github.com/basel-ay/Automated-Car-Damage-Detection (2023)

2023
[216]

Lim, Mask r-cnn for car damage detection and segmentation,https://github.com/louisyuzhe/car-damage-detector (2021)

Y. Lim, Mask r-cnn for car damage detection and segmentation,https://github.com/louisyuzhe/car-damage-detector (2021)

2021
[217]

Gupta, Automated car damage assessment,https://github.com/shubhi/car-damage-assessment (2020)

S. Gupta, Automated car damage assessment,https://github.com/shubhi/car-damage-assessment (2020)

2020
[218]

Dosovitskiy, G

A. Dosovitskiy, G. Ros, F. Codevilla, A. Lopez, V. Koltun, CARLA: An open urban driving simulator, in: Proceedings of the 1st Annual Conference on Robot Learning, 2017, pp. 1–16

2017
[219]

Kotseruba, A

I. Kotseruba, A. Rasouli, J. K. Tsotsos, Benchmark for evaluating pedestrian action prediction, in: 2021 IEEE Winter Conference on Applications of Computer Vision (WACV), 2021, pp. 1257–1267.doi:10.1109/WACV48630.2021.00130

2021
[220]

Z. Cao, G. Hidalgo, T. Simon, S.-E. Wei, Y. Sheikh, Openpose: Realtime multi-person 2d pose estimation using part affinity fields, IEEE Transactions on Pattern Analysis and Machine Intelligence 43 (1) (2021) 172–186.doi:10.1109/TPAMI.2019.2929257

2021
[221]

R. A. Guler, N. Neverova, I. Kokkinos, Densepose: Dense human pose estimation in the wild, 2018

2018
[222]

G. Zhu, S. Fan, H. Dai, E. S. L. Ho, Waymo-3dskelmo: A multi-agent 3d skeletal motion dataset for pedestrian interaction modeling in autonomous driving, in: Proceedings of the 33rd ACM International Conference on Multimedia, MM ’25, Association for Computing Machinery, New Yor...

2025
[223]

Zheng, X

J. Zheng, X. Shi, A. Gorban, J. Mao, Y. Song, C. R. Qi, T. Liu, V. Chari, A. Cornman, Y. Zhou, C. Li, D. Anguelov, Multi-modal 3d human pose estimation with 2d weak supervision in autonomous driving, arXiv (2021)

2021
[224]

Bauer, A

P. Bauer, A. Bouazizi, U. Kressel, F. B. Flohr, Weakly supervised multi-modal 3d human body pose estimation for autonomous driving, in: 2023 IEEE Intelligent Vehicles Symposium (IV), 2023, pp. 1–7.doi:10.1109/IV55152.2023.10186575

2023
[225]

S.-Y. Yu, A. V. Malawade, D. Muthirayan, P. P. Khargonekar, M. A. A. Faruque, Scene-graph augmented data-driven risk assessment of autonomous vehicle decisions, arXiv preprint arXiv:2009.06435 (2020)

2020 arXiv
[226]

University, Meaning of pedestrian in english,https://dictionary.cambridge.org/dictionary/english/pedestrian (2025)

C. University, Meaning of pedestrian in english,https://dictionary.cambridge.org/dictionary/english/pedestrian (2025)

2025
[227]

URL https://openreview.net/forum?id=OMOOO3ls6g

H.Wang,T.Li,Y.Li,L.Chen,C.Sima,Z.Liu,B.Wang,P.Jia,Y.Wang,S.Jiang,F.Wen,H.Xu,P.Luo,J.Yan,W.Zhang,H.Li,Openlane- v2: A topology reasoning benchmark for unified 3d HD mapping, in: Thirty-seventh Conference on Neural Information Processing Systems Datasets and Benchmarks Track, 20...

2023
[228]

URL https://datasetninja.com/bdd100k

D.Ninja,Visualizationtoolsforbdd100k:Images100kdataset, https://datasetninja.com/bdd100k,visitedon2025-02-01(feb2025). URL https://datasetninja.com/bdd100k
[229]

F.Yu,H.Chen,X.Wang,W.Xian,Y.Chen,F.Liu,V.Madhavan,T.Darrell,Bdd100k, https://github.com/bdd100k/bdd100k(2020)

2020
[230]

Zhang, P

S. Zhang, P. Wang, Toolkit for apolloscape dataset,https://github.com/ApolloScapeAuto/dataset-api (2022)

2022
[231]

Garnett, R

N. Garnett, R. Cohen, T. Pe’er, R. Lahav, D. Levi, 3d-lanenet: End-to-end 3d multiple lane detection, in: 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019, pp. 2921–2930.doi:10.1109/ICCV.2019.00301

2019
[232]

Y.Cai,W.Dong,Z.Liu,H.Wang,L.Chen,Homap:End-to-endvectorizedhdmapconstructionwithhigh-ordermodeling,IEEETransactions on Intelligent Vehicles 10 (5) (2025) 2987–2997.doi:10.1109/TIV.2024.3445374

2025
[233]

K. Wu, C. Yang, Z. Li, Interactionmap: Improving online vectorized hdmap construction with interaction, 2025, pp. 17176–17186.doi: 10.1109/CVPR52734.2025.01601. : Preprint submitted to Elsevier Page 83 of 85

2025
[234]

Y.Liu,T.Yuan,Y.Wang,Y.Wang,H.Zhao,Vectormapnet:end-to-endvectorizedhdmaplearning,in:Proceedingsofthe40thInternational Conference on Machine Learning, ICML’23, JMLR.org, 2023

2023
[235]

nuscenes ™ devkit, https://github.com/nutonomy/nuscenes-devkit (2020)

2020
[236]

S. Choi, J. Kim, H. Shin, J. W. Choi, Mask2map: Vectorized hd map construction using bird’s eye view segmentation masks, in: European Conference on Computer Vision, 2024

2024
[237]

Zhang, Y

Z. Zhang, Y. Zhang, X. Ding, F. Jin, X. Yue, Online vectorized hd map construction using geometry, in: Computer Vision – ECCV 2024: 18thEuropeanConference,Milan,Italy,September29–October4,2024,Proceedings,PartXLIX,Springer-Verlag,Berlin,Heidelberg,2024, p. 73–90. doi:10.1007/9...

2024 doi
[238]

Mogelmose, M

A. Mogelmose, M. M. Trivedi, T. B. Moeslund, Vision-based traffic sign detection and analysis for intelligent driver assistance systems: Perspectives and survey, IEEE Transactions on Intelligent Transportation Systems 13 (4) (2012) 1484–1497.doi:10.1109/TITS.2012. 2209421

2012 doi
[239]

Z. Zhu, D. Liang, S. Zhang, X. Huang, B. Li, S. Hu, Visualization tools for tsinghua tencent 2021 dataset,https://datasetninja.com/ tt100k-2021 (2025)

2025
[240]

citlag, Traffic-sign-recognition,https://github.com/citlag/Traffic-Sign-Recognition (2019)

2019
[241]

Vitas, M

D. Vitas, M. Tomic, M. Burul, Traffic light detection in autonomous driving systems, IEEE Consumer Electronics Magazine 9 (4) (2020) 90–96. doi:10.1109/MCE.2020.2969156

2020
[243]

Lateef, M

F. Lateef, M. Kas, Y. Ruichek, Saliency heat-map as visual attention for autonomous driving using generative adversarial network (gan), IEEE Transactions on Intelligent Transportation Systems 23 (6) (2022) 5360–5373.doi:10.1109/TITS.2021.3053178

2022
[244]

T. Deng, H. Yan, L. Qin, T. Ngo, B. S. Manjunath, How do drivers allocate their potential attention? driving fixation prediction via convolutional neural networks, IEEE Transactions on Intelligent Transportation Systems 21 (5) (2020) 2146–2154.doi:10.1109/TITS. 2019.2915540

2020
[245]

S. Gan, X. Pei, Y. Ge, Q. Wang, S. Shang, S. E. Li, B. Nie, Multisource adaption for driver attention prediction in arbitrary driving scenes, IEEE Transactions on Intelligent Transportation Systems 23 (11) (2022) 20912–20925.doi:10.1109/TITS.2022.3177640

2022
[246]

H.Kuang,K.-F.Yang,L.Chen,Y.-J.Li,L.L.H.Chan,H.Yan,Bayessaliency-basedobjectproposalgeneratorfornighttimetrafficimages, IEEE Transactions on Intelligent Transportation Systems 19 (3) (2018) 814–825.doi:10.1109/TITS.2017.2702665

2018
[247]

3225–3232.doi:10.1109/ITSC.2018.8569438

A.Tawari,P.Mallela,S.Martin,Learningtoattendtosalienttargetsindrivingvideosusingfullyconvolutionalrnn,in:201821stInternational Conference on Intelligent Transportation Systems (ITSC), 2018, pp. 3225–3232.doi:10.1109/ITSC.2018.8569438

2018
[248]

Biparva, D

M. Biparva, D. Fernández-Llorca, R. I. Gonzalo, J. K. Tsotsos, Video action recognition for lane-change classification and prediction of surrounding vehicles, IEEE Transactions on Intelligent Vehicles 7 (3) (2022) 569–578.doi:10.1109/TIV.2022.3164507

2022
[249]

Liang, J

K. Liang, J. Wang, A. Bhalerao, Lane change classification and prediction with action recognition networks, in: L. Karlinsky, T. Michaeli, K. Nishino (Eds.), Computer Vision – ECCV 2022 Workshops, Springer Nature Switzerland, Cham, 2023, pp. 617–632

2022
[251]

M. S. Kristoffersen, J. V. Dueholm, R. K. Satzoda, M. M. Trivedi, A. Møgelmose, T. B. Moeslund, Towards semantic understanding of surrounding vehicular maneuvers: A panoramic vision-based framework for real-world highway studies, in: 2016 IEEE Conference on Computer Vision and...

2016 doi
[252]

P. Shen, J. Fang, H. Yu, J. Xue, Vehicle behavior prediction by episodic-memory implanted ndt, in: 2024 IEEE International Conference on Robotics and Automation (ICRA), 2024, pp. 14177–14183.doi:10.1109/ICRA57147.2024.10610995

2024
[253]

J. Shi, J. Chen, Y. Wang, L. Sun, C. Liu, W. Xiong, T. Wo, Motion forecasting for autonomous vehicles: a survey, International Journal of Machine Learning and Cybernetics 17 (01 2026).doi:10.1007/s13042-025-02859-8

2026 doi
[254]

Teeti, S

I. Teeti, S. Khan, A. Shahbaz, A. Bradley, F. Cuzzolin, Vision-based intention and trajectory prediction in autonomous vehicles: A survey, in: L. D. Raedt (Ed.), Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, IJCAI-22, International ...

2022 doi
[255]

Singh, Trajectory-prediction with vision: A survey, in: 2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), 2023, pp

A. Singh, Trajectory-prediction with vision: A survey, in: 2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), 2023, pp. 3310–3315.doi:10.1109/ICCVW60793.2023.00356

2023
[256]

W. Zhan, L. Sun, D. Wang, H. Shi, A. Clausse, M. Naumann, J. Kümmerle, H. Königshof, C. Stiller, A. de La Fortelle, M. Tomizuka, INTERACTIONDataset:AnINTERnational,AdversarialandCooperativemoTIONDatasetinInteractiveDrivingScenarioswithSemantic Maps, arXiv:1910.03088 [cs, eess]...

1910 arXiv
[257]

K. Chen, R. Ge, H. Qiu, R. Ai-Rfou, C. R. Qi, X. Zhou, Z. Yang, S. Ettinger, P. Sun, Z. Leng, M. Mustafa, I. Bogun, W. Wang, M. Tan, D. Anguelov, Womd-lidar: Raw sensor dataset benchmark for motion forecasting, in: Proceedings of the IEEE International Conference on Robotics a...

2024
[258]

J. Gu, C. Hu, T. Zhang, X. Chen, Y. Wang, Y. Wang, H. Zhao, ViP3D: End-to-End Visual Trajectory Prediction via 3D Agent Queries , in: 2023IEEE/CVFConferenceonComputerVisionandPatternRecognition(CVPR),IEEEComputerSociety,LosAlamitos,CA,USA,2023, pp. 5496–5506. doi:10.1109/CVPR5...

2023
[259]

J. Bock, R. Krajewski, T. Moers, S. Runde, L. Vater, L. Eckstein, The ind dataset: A drone dataset of naturalistic road user trajectories at germanintersections,in:2020IEEEIntelligentVehiclesSymposium(IV),2020,pp.1929–1934. doi:10.1109/IV47402.2020.9304839

2020
[260]

Z. Zhou, J. Wang, Y.-H. Li, Y.-K. Huang, Query-centric trajectory prediction, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023, pp. 17863–17873. : Preprint submitted to Elsevier Page 84 of 85

2023
[261]

Z. Zhou, Z. Wen, J. Wang, Y.-H. Li, Y.-K. Huang, Qcnext: A next-generation framework for joint multi-agent trajectory prediction, arXiv preprint arXiv:2306.10508 (2023). : Preprint submitted to Elsevier Page 85 of 85

2023 arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

Road traffic injuries,https://www.who.int/news-room/fact-sheets/detail/road-traffic-injuries (2023)

2023

[2] [2]

B. Tian, B. T. Morris, M. Tang, Y. Liu, Y. Yao, C. Gou, D. Shen, S. Tang, Hierarchical and networked vehicle surveillance in its: A survey, IEEE Transactions on Intelligent Transportation Systems 18 (1) (2017) 25–48.doi:10.1109/TITS.2016.2552778

arXiv 2017

[3] [3]

9396–9405.doi:10.1109/CVPR.2019.00963

A.Kirillov,K.He,R.Girshick,C.Rother,P.Dollár,Panopticsegmentation,in:2019IEEE/CVFConferenceonComputerVisionandPattern Recognition (CVPR), 2019, pp. 9396–9405.doi:10.1109/CVPR.2019.00963

arXiv 2019

[4] [4]

Y.Zhou,Y.Zhang,Z.Zhao,K.Zhang,C.Gou,Towarddrivingsceneunderstanding:Aparadigmandbenchmarkdatasetforego-centrictraffic scene graph representation, IEEE Journal of Radio Frequency Identification 6 (2022) 962–967.doi:10.1109/JRFID.2022.3207017

arXiv 2022

[5] [5]

Huang, X

Y. Huang, X. Wang, Hazards prioritization with cognitive attention maps for supporting driving decision-making, IEEE Transactions on Intelligent Transportation Systems 25 (11) (2024) 16221–16234.doi:10.1109/TITS.2024.3413675

arXiv 2024

[6] [6]

L. Qin, Y. Shi, Y. He, J. Zhang, X. Zhang, Y. Li, T. Deng, H. Yan, Id-yolo: Real-time salient object detection based on the driver’s fixation region, IEEE Transactions on Intelligent Transportation Systems 23 (9) (2022) 15898–15908.doi:10.1109/TITS.2022.3146271

arXiv 2022

[7] [7]

N. Jia, Y. Sun, X. Liu, Tfgnet: Traffic salient object detection using a feature deep interaction and guidance fusion, IEEE Transactions on Intelligent Transportation Systems 25 (3) (2024) 3020–3030.doi:10.1109/TITS.2023.3293822

arXiv 2024

[8] [8]

Y. Xia, D. Zhang, J. Kim, K. Nakayama, K. Zipser, D. Whitney, Predicting driver attention in critical situations, in: Asian Conference on Computer Vision, 2017. URL https://api.semanticscholar.org/CorpusID:52019549

2017