Pith. sign in

REVIEW 2 major objections 8 minor 1 cited by

MonoSOWA: Scalable monocular 3D Object detector Without human Annotations

T0 review · 2 major / 8 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read MonoSOWA trains a 3D car detector from plain video with zero human labels, and beats prior weakly supervised methods on KITTI-360.

desk verdict A credible auto-labelling pipeline for monocular 3D detection that delivers on its main claim, with one load-bearing assumption about metric depth that deserves a sharper test. read the letter →

arxiv 2501.09481 v3 pith:JKUZWKRB submitted 2025-01-16 cs.CV cs.AIcs.LG

Jan Skvrna , Lukas Neumann This is my paper · ORCID
classification cs.CVcs.AIcs.LG
keywords monocular3Dobjectdetectionauto-labellingpseudo-LiDARweaksupervisiontemporalconsistencyLocalMotionModelCanonicalSpaceautonomousdriving
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

MonoSOWA is a method for training a monocular 3D object detector without any human annotations on the target data and without LiDAR. The idea is to replace hand-labelled 3D boxes with pseudo-labels automatically generated from video: an off-the-shelf 2D detector finds vehicles, an off-the-shelf metric depth estimator lifts each vehicle's pixels into a point cloud, and temporal aggregation across frames stabilises those noisy point clouds into 7-DOF boxes. A new Local Object Motion Model separates ego-vehicle motion from the object's own motion so that both stationary and moving cars contribute labels, and a canonical object space rescales labels to compensate for different camera focal lengths. On KITTI-360 the detector trained on these pseudo-labels beats VSRD, the previous weakly-supervised baseline, at 0.5 IoU, and the pipeline labels a frame in about 1.3 seconds versus about 15 minutes for VSRD.

What carries the argument

The load-bearing mechanism is temporal aggregation in metric pseudo-LiDAR, driven by the Local Object Motion Model (LOMM). LOMM computes per-frame median locations of each tracked instance, forms per-frame displacement vectors, and classifies the object as moving when the ratio $z = \| \mu_j \|_2 / \| \sigma_j \|_2$ exceeds a threshold while net distance also exceeds a threshold; this separates the ego-vehicle's apparent motion from the object's actual motion without any learned motion model. For stationary objects the per-frame pseudo-LiDAR clouds are concatenated into one denser cloud before box fitting; for moving objects yaw is read off the trajectory. Box fitting uses the proposed Saturated Closeness Criterion, a saturating distance that suppresses outlier points, replacing the min/max extremes of an earlier L-shape fitter with 10th/90th percentiles. Finally, every pseudo-label is rescaled by $\omega_i = f^C / f_i$ into Canonical Object Space, so that a single detector can train on and be deployed across cameras with different focal lengths.

What would settle it

Use KITTI-360's recorded LiDAR only as a validator: for the frames the pipeline auto-labels, compare each vehicle's median pseudo-LiDAR depth against the LiDAR depth for the same instance, and require the per-object depth error to stay below roughly the 0.5-IoU matching tolerance; if the depth error is systematically biased or grows with distance, the label-free accuracy result should not reproduce.

Watch

Extended reading notes

Core claim

The paper's central claim is that metric, scale-correct pseudo-LiDAR, accumulated over a driving sequence and disentangled into ego-motion and object motion, is enough supervision to train a competitive monocular 3D detector. For each vehicle, the pipeline takes a 2D instance mask from MViTv2, back-projects the pixels through a Metric3Dv2 depth map to get a per-frame point cloud, tracks the vehicle by median position, and classifies it as stationary or moving using the ratio of mean displacement to displacement jitter (LOMM). Stationary vehicles get a fused multi-frame point cloud; moving vehicles get yaw from their trajectory; a Saturated Closeness Criterion then fits an oriented 3D box, and Template Fitting Loss refines position and front/back orientation. The resulting pseudo-labels train a MonoDETR detector, with all boxes rescaled into a Canonical Object Space so that data from cameras with different focal lengths can be pooled. The paper reports that this label-free recipe reaches APBEV/AP3D@0.5 of 38.41/29.98 on KITTI-360 Easy, beating the weakly supervised VSRD (29.07/21.77), and that pre-training with pseudo-labels lets a detector fine-tuned on 25% of KITTI human labels outperform a fully supervised detector trained from scratch on 100%.

Load-bearing premise

Everything rests on the off-the-shelf monocular depth estimator producing metric, scale-correct depth for arbitrary driving scenes; if its scale is wrong, every pseudo-LiDAR point, every 3D label, and the trained detector inherit that error.

Editorial extensions

If this is right

  • Large collections of ordinary driving video, including footage from single-camera vehicles with no LiDAR, become usable as training data for monocular 3D detection.
  • The roughly 700x labelling speed-up makes auto-labelling practical at dataset scale, where the previous method's 15 minutes per frame is prohibitive.
  • With only 25% of human labels on KITTI, a detector pre-trained on MonoSOWA pseudo-labels outperforms a fully supervised detector trained on 100% of the labels; on Waymo, 15% human labels plus pseudo-labels nearly matches 100% human labels.
  • Combining pseudo-labels from KITTI and KITTI-360, two datasets with different cameras, yields accuracy comparable to training on one dataset's human labels, and exceeds human-label training in the Hard category at 0.3 IoU.
  • Because no human annotations or LiDAR are needed for a new camera setup, the same pipeline can be rerun whenever new cameras appear instead of repeating the annotation process.

Reading between the lines

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

  • Beyond the paper, if the off-the-shelf depth estimator's metric accuracy continues to improve, this pipeline should improve without any architecture change; a direct test would be replacing Metric3Dv2 with a newer metric-depth model and checking whether pseudo-label AP rises.
  • Beyond the paper, the method's components are not vehicle-specific: swapping the 2D detector for another category suggests the same temporal-aggregation recipe could auto-label pedestrians, cyclists, or other rigid objects, though the paper only evaluates cars.
  • Beyond the paper, training in canonical object space implies a detector produced this way may transfer to an unseen camera whose intrinsics are known; the paper shows cross-dataset transfer between KITTI and KITTI-360 but does not test truly novel camera geometries.
  • The authors speculate that fusing sparse low-quality LiDAR with dense pseudo-LiDAR could correct depth drift; if that speculation holds, it offers a practical middle path between fully label-free and LiDAR-assisted labelling.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 8 minor

Summary. The paper presents MonoSOWA, an automatic pipeline that generates 3D vehicle labels from monocular video without target-domain human annotations. The pipeline lifts per-frame metric depth predictions (Metric3Dv2) into pseudo-LiDAR, segments vehicles with a COCO-pretrained 2D instance segmentation model (MViTv2-H), tracks objects across frames, classifies each instance as stationary or moving using a proposed Local Object Motion Model (LOMM), and aggregates observations to estimate 3D position, dimensions, and yaw. A Saturated Closeness Criterion is introduced for L-shape fitting, and labels are transformed into a Canonical Object Space (COS) to mix data from cameras with different focal lengths. A MonoDETR detector trained on the resulting pseudo-labels is evaluated on KITTI-360, Waymo, and KITTI, where it outperforms prior weakly-supervised monocular methods at 0.5 IoU, serves as an effective pre-training initialization for fine-tuning with small amounts of human labels, and benefits from combining pseudo-labels across datasets. The auto-labelling pipeline runs in roughly 1.3 seconds per frame, about 700x faster than VSRD.

Significance. The significance of the work is substantial if the results hold. It demonstrates a practical, fully automatic alternative to human 3D annotation for monocular detection, with a fast pipeline and the ability to aggregate data from multiple camera setups. The experiments on three public datasets, the cross-dataset combination, and the pre-training results are valuable. The authors also release code and models. The main caveats are that the pipeline relies on off-the-shelf models whose training used human supervision (so the 'without human annotations' claim must be carefully scoped), and the generalization across new cameras hinges on the metric accuracy of the monocular depth estimator, which is not directly validated on the target datasets. These points do not invalidate the headline results, but they need to be addressed for the scalability claim.

major comments (2)
  1. [Section III-A, Eq. (1); Supplementary Table XII] The entire auto-labelling pipeline assumes Metric3Dv2 provides metric, scale-correct depth for arbitrary driving scenes. The paper does not evaluate depth scale directly on the three target datasets. The only quantitative comparison is Table XII on KITTI, where replacing pseudo-LiDAR with real LiDAR in the same pipeline improves APBEV@0.5 Easy from 59.76 to 64.50, indicating non-trivial depth errors. Because the claimed scalability to new camera setups and datasets rests on the depth estimator's metric accuracy, please provide a direct depth-scale evaluation (e.g., median absolute relative error, or scale bias against LiDAR/ground-truth depth) on KITTI-360 and Waymo, or explicitly discuss why the end-to-end detection AP is sufficient evidence.
  2. [Section IV-A, Table X] The text states that when identical inputs are used, our method significantly outperforms VSRD in all metrics (referring to Table X). This is contradicted by Table X: with human-annotated masks, the AP3D@0.3 Easy value for the proposed method is 49.03, below VSRD's 50.86. The corresponding BEV value (59.41 vs 58.40) and the Hard values do improve, but 'all metrics' is incorrect. Please revise this sentence and reassess the explanation for the 0.3 IoU gap in Table I.
minor comments (8)
  1. [Title / Abstract] The title and the phrase 'without human Annotations' overclaim because the pipeline uses a COCO-pretrained 2D detector and a depth estimator that were trained with human-annotated data. The abstract's qualifier 'domain-specific human annotations' is accurate, but the title and the first contribution sentence should be scoped to 'target-domain human annotations' to avoid misleading readers.
  2. [Section III-A, Eq. (1)] The variable ranges are swapped: 'u ∈ [0, h]' should be 'u ∈ [0, w]' and 'v ∈ [0, h]' (image width and height), and the text says 'h and w denotes image height and width'.
  3. [Section III-D, Eq. (9)] The third formula should read zC = z · ωi, not zC = y · ωi.
  4. [Section III-B, Eq. (4)] The definition of σj differs from the usual standard deviation by a factor of 1/√2. Please clarify whether this is intentional and how it affects the z-ratio threshold.
  5. [Section IV-B, hyperparameters] The thresholds Tz and Tm in the LOMM are fixed across datasets, but no ablation is reported for them. Please add a sensitivity analysis or justify the chosen values.
  6. [Section IV-A, KITTI-360 paragraph] The phrase 'when identical inputs are used, our method significantly outperforms VSRD in all metrics' should be corrected as described in the major comment.
  7. [General] The experiments do not report error bars or multiple seeds; given the small gaps in some comparisons (e.g., Table X AP3D@0.3 Easy), please indicate the variance of the reported AP values.
  8. [Typos] The header/title contains 'MonoSOW A' with an unwanted space; please correct the spacing throughout.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: pseudo-labels are produced by a fixed pipeline from independent off-the-shelf models, and the detector is evaluated on held-out human-annotated splits.

full rationale

The derivation is self-contained. Pseudo-labels are generated by a fixed pipeline (Eq. 1 back-projection; Eq. 3 tracking; LOMM Eqs. 4-6; COS Eqs. 8-9) whose inputs are off-the-shelf models (Metric3Dv2, MViTv2) and ego-motion; no equation in the paper defines a target quantity in terms of the quantity it is claimed to predict. The detector (MonoDETR) is trained on pseudo-labels and evaluated on held-out human-annotated splits (KITTI-360 test, Waymo validation, KITTI validation), so the headline AP numbers are not fitted inputs renamed as predictions. Hyperparameters (Tz, Tm, alpha, fc) are set by ablations on KITTI training pseudo-label quality, not by fitting to the final test metrics, and the same settings are used across all three datasets. The only overlapping-author citations ([30] TCC-Det for Template Fitting Loss; [29] for point-cloud extraction inspiration) supply component losses or design inspiration, not the uniqueness of the method or the validity of the central claim, so they are not load-bearing. The depth-scale dependence on Metric3Dv2 is an external-assumption risk (correctness), not a circularity.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The method is an engineering pipeline composed of off-the-shelf models and hand-set thresholds. The free parameters are chosen via ablations or experience rather than derived from theory, but they are not fitted to test-set results. No new physical entities or forces are introduced.

free parameters (5)
  • Tz (stationary/moving ratio threshold) = 0.2
    Threshold on z-score in Eq. (6) to classify instances as moving; chosen by hand, not derived from first principles.
  • Tm (minimum net distance threshold) = 5 meters
    Minimum net displacement for an instance to be classified as moving; hand-set and dataset-independent.
  • alpha (Saturated Closeness steepness) = 10
    Steepness parameter in Algorithm 1; ablated in Table XIII, chosen for best AP.
  • Canonical focal length fc = 750
    Canonical focal length for COS transformation; ablated in Table XIV, chosen as close to actual camera focal lengths.
  • Aggregation window = 100 frames
    Number of frames aggregated in auto-labelling; ablated in Table XI.
assumptions (5)
  • domain assumption Metric3Dv2 produces metric, scale-correct depth estimates for driving scenes.
    The entire pseudo-LiDAR generation in Section III-A assumes depth values are in world meters; no independent scale verification is provided.
  • domain assumption Ego-vehicle motion from GPS/IMU is available and approximately accurate.
    Section III states approximate ego-motion data is required to transform point clouds to the reference frame; inaccuracies propagate into labels.
  • domain assumption Cars are rigid objects whose motion between frames can be modeled as translation plus yaw.
    The tracking motion model (Eq. 3) and yaw estimation (Eq. 7) assume rigid-body motion with negligible lateral slip or roll.
  • domain assumption Camera intrinsics and extrinsics are known for each sequence.
    The unprojection in Eq. (1) and the COS transformation (Eq. 8) require known focal length and principal point; the paper states this is assumed.
  • domain assumption The 2D instance detector (MViTv2 trained on MS-COCO) generalizes to driving scenes.
    The pipeline relies on COCO-trained detection masks to segment vehicles; domain shift to autonomous driving scenes is not explicitly measured.

how reviews work

0 comments
Cite this review

Pith. "Pith review of MonoSOWA: Scalable monocular 3D Object detector Without human Annotations." pith.science (2026). https://pith.science/paper/JKUZWKRB

@misc{pith2026250109481,
  author       = {Pith},
  title        = {Pith review of: MonoSOWA: Scalable monocular 3D Object detector Without human Annotations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JKUZWKRB}},
  note         = {Machine review of arXiv:2501.09481}
}
read the original abstract

Inferring object 3D position and orientation from a single RGB camera is a foundational task in computer vision with many important applications. Traditionally, 3D object detection methods are trained in a fully-supervised setup, requiring LiDAR and vast amounts of human annotations, which are laborious, costly, and do not scale well with the ever-increasing amounts of data being captured. We present a novel method to train a 3D object detector from a single RGB camera without domain-specific human annotations, making orders of magnitude more data available for training. The method uses newly proposed Local Object Motion Model to disentangle object movement source between subsequent frames, is approximately 700 times faster than previous work and compensates camera focal length differences to aggregate multiple datasets. The method is evaluated on three public datasets, where despite using no human labels, it outperforms prior work by a significant margin. It also shows its versatility as a pre-training tool for fully-supervised training and shows that combining pseudo-labels from multiple datasets can achieve comparable accuracy to using human labels from a single dataset. The source code and model are available at https://github.com/jskvrna/MonoSOWA.

Figures

Figures reproduced from arXiv: 2501.09481 by the authors.

Figure 1
Figure 1. Traditional pipelines (a) vs. the proposed pipeline (b) for monocular 3D object detection. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Our auto-labelling pipeline creates 3D labels using only images and ego-motion data as inputs (and known camera [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Combining instance segmentation in image (top) with [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: The point cloud of a significantly occluded vehicle (a) [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: Generated 3D labels are transformed into Canonical [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Example of 3D car detections by MonoDETR [ [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Qualitative analysis of MonoDETR [20] trained using our method without using human annotations on KITTI [10] and KITTI-360 [42] . White 3D bounding boxes are predictions, the number inside/near the bounding box is the confidence of each prediction. 13 [PITH_FULL_IMAGE…
Figure 8
Figure 8. Figure 8: Qualitative analysis of MonoDETR [20] trained using our method without using human annotations on Waymo Open Dataset [44]. Colourful 3D bounding boxes are predictions of our model. 14 [PITH_FULL_IMAGE:figures/full_fig_p014_8.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. PLOT: Pseudo-Labeling via Object Tracking for Monocular 3D Object Detection

    cs.CV 2025-07 conditional novelty 6.0 of 10

    PLOT generates 3D bounding-box labels for objects from monocular video alone, using point tracking, estimated camera motion, and temporal fusion of pseudo-LiDAR, without auxiliary sensors or model training.

Reference graph

Works this paper leans on

49 extracted references · 45 canonical work pages · cited by 1 Pith paper

  1. [1]

    Weakm3d: Towards weakly supervised monocular 3d object detection,

    L. Peng, S. Yan, B. Wu, Z. Yang, X. He, and D. Cai, “Weakm3d: Towards weakly supervised monocular 3d object detection,” in Inter- national Conference on Learning Representations , 2022. 1, 6

    work page 2022

  2. [2]

    Autolabeling 3d objects with differentiable rendering of sdf shape priors,

    S. Zakharov, W. Kehl, A. Bhargava, and A. Gaidon, “Autolabeling 3d objects with differentiable rendering of sdf shape priors,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, pp. 12224–12233, 2020. 1, 2, 6

    work page 2020

  3. [3]

    Weakly supervised monocular 3d object detection using multi-view projection and direction consistency,

    R. Tao, W. Han, Z. Qiu, C.-Z. Xu, and J. Shen, “Weakly supervised monocular 3d object detection using multi-view projection and direction consistency,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 17482–17492, 2023. 2

    work page 2023

  4. [4]

    Vsrd: Instance-aware volumetric silhouette rendering for weakly supervised 3d object detection,

    Z. Liu, H. Sakuma, and M. Okutomi, “Vsrd: Instance-aware volumetric silhouette rendering for weakly supervised 3d object detection,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 17354–17363, 2024. 2, 6, 7, 11

    work page 2024

  5. [5]

    Metric3d: Towards zero-shot metric 3d prediction from a sin- gle image,

    W. Yin, C. Zhang, H. Chen, Z. Cai, G. Yu, K. Wang, X. Chen, and C. Shen, “Metric3d: Towards zero-shot metric 3d prediction from a sin- gle image,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , pp. 9043–9053, 2023. 2, 3, 5

    work page 2023

  6. [6]

    Omni3d: A large benchmark and model for 3d object detection in the wild,

    G. Brazil, A. Kumar, J. Straub, N. Ravi, J. Johnson, and G. Gkioxari, “Omni3d: A large benchmark and model for 3d object detection in the wild,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pp. 13154–13164, 2023. 2, 5

    work page 2023

  7. [7]

    Towards robust monocular depth estimation: Mixing datasets for zero-shot cross- dataset transfer,

    R. Ranftl, K. Lasinger, D. Hafner, K. Schindler, and V . Koltun, “Towards robust monocular depth estimation: Mixing datasets for zero-shot cross- dataset transfer,” IEEE transactions on pattern analysis and machine intelligence, vol. 44, no. 3, pp. 1623–1637, 2020. 2 8

    work page 2020

  8. [8]

    Zoedepth: Zero-shot transfer by combining relative and metric depth,

    S. F. Bhat, R. Birkl, D. Wofk, P. Wonka, and M. M ¨uller, “Zoedepth: Zero-shot transfer by combining relative and metric depth,” arXiv preprint arXiv:2302.12288, 2023. 2

    arXiv 2023

Show all 49 references
  1. [9]

    Depth anything: Unleashing the power of large-scale unlabeled data,

    L. Yang, B. Kang, Z. Huang, X. Xu, J. Feng, and H. Zhao, “Depth anything: Unleashing the power of large-scale unlabeled data,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 10371–10381, 2024. 2

    2024

  2. [10]

    Vision meets robotics: The kitti dataset,

    A. Geiger, P. Lenz, C. Stiller, and R. Urtasun, “Vision meets robotics: The kitti dataset,” The International Journal of Robotics Research , vol. 32, no. 11, pp. 1231–1237, 2013. 2, 6, 7, 8, 11, 12, 13

    2013

  3. [11]

    Metric3d v2: A versatile monocular geometric foundation model for zero-shot metric depth and surface normal estimation,

    M. Hu, W. Yin, C. Zhang, Z. Cai, X. Long, H. Chen, K. Wang, G. Yu, C. Shen, and S. Shen, “Metric3d v2: A versatile monocular geometric foundation model for zero-shot metric depth and surface normal estimation,” arXiv preprint arXiv:2404.15506 , 2024. 2, 6, 11

    2024 arXiv

  4. [12]

    Dinov2: Learning robust visual features without supervision,

    M. Oquab, T. Darcet, T. Moutakanni, H. V o, M. Szafraniec, V . Khalidov, P. Fernandez, D. Haziza, F. Massa, A. El-Nouby, et al. , “Dinov2: Learning robust visual features without supervision,” arXiv preprint arXiv:2304.07193, 2023. 2

    2023 arXiv

  5. [13]

    A convnet for the 2020s,

    Z. Liu, H. Mao, C.-Y . Wu, C. Feichtenhofer, T. Darrell, and S. Xie, “A convnet for the 2020s,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pp. 11976–11986, 2022. 2

    2022

  6. [14]

    Vision transformers for dense prediction,

    R. Ranftl, A. Bochkovskiy, and V . Koltun, “Vision transformers for dense prediction,” in Proceedings of the IEEE/CVF international conference on computer vision , pp. 12179–12188, 2021. 2

    2021

  7. [15]

    Monocular 3d object detection with pseudo- lidar point cloud,

    X. Weng and K. Kitani, “Monocular 3d object detection with pseudo- lidar point cloud,” in Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops , pp. 0–0, 2019. 2

    2019

  8. [16]

    Pseudo-lidar from visual depth estimation: Bridging the gap in 3d object detection for autonomous driving,

    Y . Wang, W.-L. Chao, D. Garg, B. Hariharan, M. Campbell, and K. Q. Weinberger, “Pseudo-lidar from visual depth estimation: Bridging the gap in 3d object detection for autonomous driving,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , p...

    2019

  9. [17]

    Accurate monocular 3d object detection via color-embedded 3d reconstruction for autonomous driving,

    X. Ma, Z. Wang, H. Li, P. Zhang, W. Ouyang, and X. Fan, “Accurate monocular 3d object detection via color-embedded 3d reconstruction for autonomous driving,” in Proceedings of the IEEE/CVF international conference on computer vision , pp. 6851–6860, 2019. 2

    2019

  10. [18]

    Smoke: Single-stage monocular 3d object detection via keypoint estimation,

    Z. Liu, Z. Wu, and R. T ´oth, “Smoke: Single-stage monocular 3d object detection via keypoint estimation,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops , pp. 996–997, 2020. 2

    2020

  11. [19]

    Monodtr: Monocular 3d object detection with depth-aware transformer,

    K.-C. Huang, T.-H. Wu, H.-T. Su, and W. H. Hsu, “Monodtr: Monocular 3d object detection with depth-aware transformer,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pp. 4012–4021, 2022. 2

    2022

  12. [20]

    Monodetr: Depth-guided transformer for monocular 3d object detection,

    R. Zhang, H. Qiu, T. Wang, Z. Guo, Z. Cui, Y . Qiao, H. Li, and P. Gao, “Monodetr: Depth-guided transformer for monocular 3d object detection,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , pp. 9155–9166, 2023. 2, 6, 7, 8, 11, 12, 13, 14

    2023

  13. [21]

    Attention is all you need,

    A. Vaswani, “Attention is all you need,” Advances in Neural Information Processing Systems, 2017. 2

    2017

  14. [22]

    Ssd-monodetr: Supervised scale-aware deformable transformer for monocular 3d object detection,

    X. He, F. Yang, K. Yang, J. Lin, H. Fu, M. Wang, J. Yuan, and Z. Li, “Ssd-monodetr: Supervised scale-aware deformable transformer for monocular 3d object detection,” IEEE Transactions on Intelligent Vehicles, 2023. 2

    2023

  15. [23]

    Monoatt: Online monocular 3d object detection with adaptive token transformer,

    Y . Zhou, H. Zhu, Q. Liu, S. Chang, and M. Guo, “Monoatt: Online monocular 3d object detection with adaptive token transformer,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 17493–17503, 2023. 2

    2023

  16. [24]

    Objects are different: Flexible monocular 3d object detection,

    Y . Zhang, J. Lu, and J. Zhou, “Objects are different: Flexible monocular 3d object detection,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 3289–3298, 2021. 2, 6

    2021

  17. [25]

    Monocd: Monocular 3d object detection with complementary depths,

    L. Yan, P. Yan, S. Xiong, X. Xiang, and Y . Tan, “Monocd: Monocular 3d object detection with complementary depths,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 10248–10257, 2024. 2

    2024

  18. [26]

    Weakly supervised 3d object detection from point clouds,

    Z. Qin, J. Wang, and Y . Lu, “Weakly supervised 3d object detection from point clouds,” in Proceedings of the 28th ACM International Conference on Multimedia, pp. 4144–4152, 2020. 2

    2020

  19. [27]

    Weakly supervised 3d object detection from lidar point cloud,

    Q. Meng, W. Wang, T. Zhou, J. Shen, L. Van Gool, and D. Dai, “Weakly supervised 3d object detection from lidar point cloud,” in ECCV, 2020. 2

    2020

  20. [28]

    Towards a weakly supervised framework for 3d point cloud object detection and annotation,

    Q. Meng, W. Wang, T. Zhou, J. Shen, Y . Jia, and L. Van Gool, “Towards a weakly supervised framework for 3d point cloud object detection and annotation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 8, pp. 4454–4468, 2021. 2

    2021

  21. [29]

    Lifting 2D object locations to 3D by discounting LiDAR outliers across objects and views,

    R. McCraith, E. Insafutdinov, L. Neumann, and A. Vedaldi, “Lifting 2D object locations to 3D by discounting LiDAR outliers across objects and views,” in 2022 International Conference on Robotics and Automation (ICRA), pp. 2411–2418, 2022. 2, 3, 6

    2022

  22. [30]

    TCC-Det: Temporarily consistent cues for weakly-supervised 3D detection,

    J. Skvrna and L. Neumann, “TCC-Det: Temporarily consistent cues for weakly-supervised 3D detection,” in European Conference on Computer Vision, pp. 129–145, Springer, 2024. 2, 4, 5, 6

    2024

  23. [31]

    Weakly supervised 3d object detection via multi-level visual guidance,

    K.-C. Huang, Y .-H. Tsai, and M.-H. Yang, “Weakly supervised 3d object detection via multi-level visual guidance,” in ECCV, 2024. 2

    2024

  24. [32]

    Open-set 3d object detection,

    J. Cen, P. Yun, J. Cai, M. Y . Wang, and M. Liu, “Open-set 3d object detection,” in 2021 International conference on 3D vision (3DV) , pp. 869–878, IEEE, 2021. 2

    2021

  25. [33]

    4d unsupervised object dis- covery,

    Y . Wang, Y . Chen, and Z.-X. ZHANG, “4d unsupervised object dis- covery,” Advances in Neural Information Processing Systems , vol. 35, pp. 35563–35575, 2022. 2

    2022

  26. [34]

    Motion inspired unsupervised perception and prediction in autonomous driving,

    M. Najibi, J. Ji, Y . Zhou, C. R. Qi, X. Yan, S. Ettinger, and D. Anguelov, “Motion inspired unsupervised perception and prediction in autonomous driving,” in European Conference on Computer Vision , pp. 424–443, Springer, 2022. 2

    2022

  27. [35]

    Learning to detect mobile objects from lidar scans without labels,

    Y . You, K. Luo, C. P. Phoo, W.-L. Chao, W. Sun, B. Hariharan, M. Campbell, and K. Q. Weinberger, “Learning to detect mobile objects from lidar scans without labels,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 1130– 1140, 2022. 2

    2022

  28. [36]

    Reward finetuning for faster and more accurate unsupervised object discovery,

    K. Luo, Z. Liu, X. Chen, Y . You, S. Benaim, C. P. Phoo, M. Campbell, W. Sun, B. Hariharan, and K. Q. Weinberger, “Reward finetuning for faster and more accurate unsupervised object discovery,” Advances in Neural Information Processing Systems, vol. 36, pp. 13250–13266, 2023. 2

    2023

  29. [37]

    Towards unsupervised object detection from lidar point clouds,

    L. Zhang, A. J. Yang, Y . Xiong, S. Casas, B. Yang, M. Ren, and R. Urtasun, “Towards unsupervised object detection from lidar point clouds,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 9317–9328, 2023. 2

    2023

  30. [38]

    Semoli: What moves together belongs together,

    J. Seidenschwarz, A. Osep, F. Ferroni, S. Lucey, and L. Leal-Taix ´e, “Semoli: What moves together belongs together,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pp. 14685–14694, 2024. 2

    2024

  31. [39]

    Liso: Lidar-only self- supervised 3d object detection,

    S. A. Baur, F. Moosmann, and A. Geiger, “Liso: Lidar-only self- supervised 3d object detection,” in European Conference on Computer Vision, pp. 253–270, Springer, 2024. 2

    2024

  32. [40]

    Mvitv2: Improved multiscale vision transformers for classification and detection,

    Y . Li, C.-Y . Wu, H. Fan, K. Mangalam, B. Xiong, J. Malik, and C. Feichtenhofer, “Mvitv2: Improved multiscale vision transformers for classification and detection,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pp. 4804–4814,

  33. [41]

    Efficient l-shape fitting for vehicle detection using laser scanners,

    X. Zhang, W. Xu, C. Dong, and J. M. Dolan, “Efficient l-shape fitting for vehicle detection using laser scanners,” in 2017 IEEE Intelligent Vehicles Symposium (IV), pp. 54–59, IEEE, 2017. 4

    2017

  34. [42]

    Kitti-360: A novel dataset and bench- marks for urban scene understanding in 2d and 3d,

    Y . Liao, J. Xie, and A. Geiger, “Kitti-360: A novel dataset and bench- marks for urban scene understanding in 2d and 3d,” IEEE Transactions on Pattern Analysis and Machine Intelligence , vol. 45, no. 3, pp. 3292– 3310, 2022. 6, 7, 8, 11, 12, 13

    2022

  35. [43]

    Deviant: Depth equivariant network for monocular 3d object detection,

    A. Kumar, G. Brazil, E. Corona, A. Parchami, and X. Liu, “Deviant: Depth equivariant network for monocular 3d object detection,” in European Conference on Computer Vision, pp. 664–683, Springer, 2022. 7

    2022

  36. [44]

    Scalability in perception for autonomous driving: Waymo open dataset,

    P. Sun, H. Kretzschmar, X. Dotiwalla, A. Chouard, V . Patnaik, P. Tsui, J. Guo, Y . Zhou, Y . Chai, B. Caine, V . Vasudevan, W. Han, J. Ngiam, H. Zhao, A. Timofeev, S. Ettinger, M. Krivokon, A. Gao, A. Joshi, Y . Zhang, J. Shlens, Z. Chen, and D. Anguelov, “Scalability in perc...

    2020

  37. [45]

    Fgr: Frustum-aware geometric reasoning for weakly supervised 3d vehicle detection,

    Y . Wei, S. Su, J. Lu, and J. Zhou, “Fgr: Frustum-aware geometric reasoning for weakly supervised 3d vehicle detection,” in 2021 IEEE In- ternational Conference on Robotics and Automation (ICRA) , pp. 4348– 4354, IEEE, 2021. 6

    2021

  38. [46]

    Detectron2

    Y . Wu, A. Kirillov, F. Massa, W.-Y . Lo, and R. Girshick, “Detectron2.” https://github.com/facebookresearch/detectron2, 2019. 6

    2019

  39. [47]

    Microsoft COCO: Common objects in context,

    T.-Y . Lin, M. Maire, S. Belongie, J. Hays, P. Perona, D. Ramanan, P. Doll ´ar, and C. L. Zitnick, “Microsoft COCO: Common objects in context,” in Computer Vision–ECCV 2014: 13th European Conference, 9 Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13 , pp. 740...

    2014

  40. [48]

    Decoupled weight decay regularization,

    I. Loshchilov, “Decoupled weight decay regularization,” arXiv preprint arXiv:1711.05101, 2017. 6

    2017 arXiv

  41. [49]

    Efficient variants of the icp algorithm,

    S. Rusinkiewicz and M. Levoy, “Efficient variants of the icp algorithm,” in Proceedings Third International Conference on 3-D Digital Imaging and Modeling, pp. 145–152, 2001. 11 10 MonoSOW A: Scalable monocular 3D Object detector Without human Annotations Supplementary Materia...

    2001

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.