Pith. sign in

REVIEW 2 major objections 5 minor 34 references

MORDA: A Synthetic Dataset to Facilitate Adaptation of Object Detectors to Unseen Real-target Domain While Preserving Performance on Real-source Domain

T0 review · 2 major / 5 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 A synthetic-fusion dataset built from South Korean digital twins and nuScenes-style sensors lets object detectors generalize to an unseen real target without losing source-domain performance.

desk verdict A genuinely new synthetic dataset that improves detection on an unseen target domain, but the paper never isolates whether the Korea-specific digital twins or just extra synthetic data drive the gain. read the letter →

arxiv 2501.04950 v3 pith:4SLGF3ET submitted 2025-01-09 cs.CV

Hojun Lim , Heecheol Yoo , Jinwoo Lee , Seungmin Jeon , Hyeongseok Jeon This is my paper · ORCID
classification cs.CV
keywords MORDAdatasetsyntheticdrivingdomainadaptationdigitaltwinobjectdetectionautonomousnuScenessim-to-realtransfer
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

The paper tries to show that a synthetic driving dataset can prepare object detectors for a real region they have never seen, without forgetting the region they were trained on. It builds MORDA by putting nuScenes-style sensors, annotation rules, and object classes into a simulator stocked with digital-twin maps of several South Korean regions. Detectors trained on plain nuScenes plus MORDA, with no domain-adaptation trick, score substantially better on the unseen real South Korean AI-Hub dataset across 2D camera and 3D LiDAR detectors, while nuScenes scores stay flat or rise. The authors also report that MORDA beats other synthetic datasets for 2D detection with far fewer frames. If the result holds, simulating the target region plus matching the source sensor and labeling pipeline could be a cheap substitute for recollecting and relabeling data in every new deployment country.

What carries the argument

The load-bearing object is the synthetic-fusion domain $D^{Src+Trg}_{Syn}$, a simulated world in which the target region's digital-twin maps are populated with the source dataset's sensor suite and labeling scheme. In MORDA this means nuScenes' six-camera and 32-beam LiDAR geometry, ten detection classes, and 3D-box conventions, including articulated bus and truck-trailer rules, are reproduced inside simulator maps of one highway and three urban South Korean locations, with static and dynamic scenes recorded at 20 Hz. The dataset is then converted to nuScenes format so that existing detectors can ingest it by simple concatenation. This pairing of target geography with source sensor and label fidelity is what the paper claims carries the adaptation, and it is what distinguishes MORDA from synthetic datasets built without a specific real target in mind.

What would settle it

Train the same detectors on nuScenes plus synthetic data generated with identical sensors and annotation rules but digital-twin maps of a different country; if the AI-Hub mAP gain is unchanged, the target-specific Korean maps are not what carries the result. A complementary check is to evaluate the same models on a second unseen target city outside Korea, where MORDA should confer no advantage if the effect is genuinely target-specific.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a synthetic-fusion domain, a simulator world that blends the target region's geography with the source dataset's sensing and labeling protocol, transfers to an unseen real target domain. Using nuScenes as source and South Korea as target, the paper creates MORDA, the Mixture Of Real-domain characteristics for synthetic-data-assisted Domain Adaptation: 87 scenes, 37K frames, 1.6M 3D boxes, six 1600x900 cameras and one 32-beam LiDAR following nuScenes, and digital-twin maps of one highway and three urban South Korean areas. Training simple detectors on nuScenes plus MORDA raises AI-Hub mAP from 13.35 to 19.7 for Faster R-CNN and from 7.20 to 22.93 for CenterPoint with a 0.1m voxel backbone, while nuScenes mAP improves slightly; similar gains appear for PointPillars and SSN. The paper interprets this as evidence that the simulator provides a preview of the target domain and acts as a regularizer against generalization failure.

Load-bearing premise

The load-bearing premise is that the digital-twin maps of South Korea used to build MORDA faithfully reproduce the visual and geometric properties of the real target that matter for object detection, even though the real AI-Hub data were not used to construct them.

Editorial extensions

If this is right

  • Adding MORDA to nuScenes training improves 2D and 3D detection on the unseen AI-Hub target across all tested architectures, with the largest gains where baseline models collapse under domain shift.
  • Source-domain performance is retained or slightly improved, so the method does not trade away the original deployment domain.
  • MORDA outperforms VKITTI2, SYNTHIA-AL, and SHIFT for 2D detection with fewer frames, suggesting target-matched synthetic data is more efficient than larger generic synthetic data.
  • Because only simple concatenation is used, combining MORDA with existing unsupervised domain adaptation methods is a direct next step that could yield further gains.
  • The recipe, if it generalizes, gives AV developers a data-acquisition path to a new region that avoids dispatching a sensor vehicle there for labeling.

Reading between the lines

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

  • If the target-specific digital-twin maps are the active ingredient, the same pipeline should transfer to other source-target pairs; building a non-Korean twin with identical sensors and labels and comparing AI-Hub gains would test this directly.
  • The AI-Hub evaluation uses converted labels and pseudo-2D boxes derived from panoptic segmentation, so part of the reported gain could reflect alignment with the converted label distribution rather than raw perception quality; a small re-annotation study would separate these.
  • MORDA's heavy representation of rare classes such as trucks, construction vehicles, and trailers may be doing much of the work; an ablation that matches nuScenes' class distribution in the simulator would reveal whether class balancing or geographic fidelity drives the gain.
  • The method suggests a general principle: synthetic data's value for domain adaptation comes from matching the source annotation protocol and the target environment simultaneously, not from photorealism alone.
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.

Referee Report

2 major / 5 minor

Summary. The paper introduces MORDA, a synthetic driving dataset built by combining digital-twin maps of South Korea with a reproduction of the nuScenes sensor suite, object categories, and 3D bounding-box annotation rules. The authors train 2D camera-based and 3D LiDAR-based detectors on nuScenes alone or nuScenes plus MORDA, then evaluate on the nuScenes validation set and on the AI-Hub dataset collected in South Korea, which is never used for training. They report that adding MORDA improves mean average precision on AI-Hub across all tested detectors (e.g., +6.35 mAP for Faster-RCNN, +5.71 for SSN, +22.17 for CenterPoint-Voxel-0.075m) while nuScenes mAP/NDS are maintained or slightly improved. A comparison with VKITTI2, SYNTHIA-AL, and SHIFT is performed only on nuScenes validation.

Significance. If the central claim survives, the paper offers a cost-effective alternative to collecting and labeling real data in a new deployment region, and MORDA is a large, openly documented synthetic dataset with annotations for multiple tasks. The strengths are the breadth of detectors (2D and 3D, multiple architectures), the faithful reproduction of nuScenes sensor geometry and label conventions, and the clear demonstration that adding synthetic data can rescue detectors from near-zero performance under a large domain shift (CenterPoint). The main weakness is that the target-domain gains are not shown to require the Korean digital twins, because no generic synthetic dataset is evaluated on AI-Hub.

major comments (2)
  1. [Section V-C, Table IV; Section VI] The paper's central claim is that MORDA's value comes from fusing nuScenes characteristics with Korean target-domain geography, but the only synthetic-dataset comparison is conducted on nuScenes validation, not on AI-Hub. Table IV shows that VKITTI2, SYNTHIA-AL, and SHIFT give mAP gains of -0.3, +0.1, and +0.4 on nuScenes, yet none of these datasets is tested on AI-Hub. Therefore, the large AI-Hub improvements in Table II (e.g., +18.16 and +22.17 mAP for CenterPoint voxel models) could in principle be obtained by any additional synthetic data that regularizes the detector, without any contribution from the Korean digital-twin maps. The authors themselves defer attribution to future work in Section VI ('identify which characteristics implemented in MORDA contributed to performance stability on DT_rg'). Please add a control experiment that trains on nuScenes plus a non-Korean synthetic dataset (or synthetic data generated from non-Korean maps) and evaluates on AI-Hub, to isolate the effect of the target-specific geography.
  2. [Section V-A.4 and Table II] All results are from a single training run per condition, and no error bars or significance tests are provided. The nuScenes 'retained or slightly enhanced' claim relies on small differences (e.g., +0.5 mAP for Faster-RCNN, +0.51 for PointPillars), which could be within run-to-run variance. Please report mean and standard deviation over at least three seeds for the main comparisons, or otherwise justify why single runs are sufficient.
minor comments (5)
  1. [Table I] OPV2V is a vehicle-to-vehicle communication dataset rather than a single-vehicle driving dataset; the comparison table would benefit from a note clarifying this distinction.
  2. [Section V-A.2] The sentence 'we use the implementation of the mentioned networks in MMDetection3D' should say 'we use the implementations'.
  3. [Fig. 4] The red dotted ellipse mentioned in the text is not clearly visible in grayscale; please increase its contrast or add a zoomed inset.
  4. [Throughout] The term 'A Vs' is inconsistently spaced; consider using 'AVs' for uniformity.
  5. [Dataset access] The dataset is announced via a GitLab page; please state the license and whether the generation code and assets will be released publicly.

Circularity Check

1 steps flagged · score 4.0 of 10

Partial circularity: MORDA's target-preview ability is built into the dataset by construction, but the measured detector gains are empirical.

  1. self definitional [Section III-C and Section V-B.1; cf. Section VI]
    "we employ four digital-twin maps of South Korea available in MORAI SIM ... each digital-twin map was constructed upon HD map data, geographical survey data, and 3D building data collected from each counterparting region in South Korea ... AI-Hub is opted to represent DT_rg_Real given (1) its sensor data are collected from the real-target domain of this paper, South Korea where digital twin maps in MORDA are originated from."

    The synthetic dataset is defined as a virtual replica of the target region's geography, so the paper's claim that MORDA 'previews' or 'provides simulated real-world experiences' of DT_rg_Real holds by construction rather than by empirical demonstration. The non-trivial empirical finding is the detector mAP gain on AI-Hub, but the paper does not isolate whether that gain comes from the Korean digital-twin content: Table IV compares other synthetic datasets only on nuScenes, never on AI-Hub, and Section VI defers this attribution to future work. Hence the target-specific explanation is partly circular and untested, while the measured improvements themselves are not forced.

full rationale

The paper's derivation chain is mostly empirical: MORDA is generated by reproducing nuScenes' sensor and labeling setup inside MORAI SIM and placing those sensors on Korean digital-twin maps; detectors are then trained on nuScenes plus MORDA and evaluated on AI-Hub without using AI-Hub sensor data during training. No parameter is fitted to AI-Hub and no detection metric is forced by an equation, so the measured mAP gains are not circular in the statistical sense. The self-citation to MORAI SIM [30] is a normal tool citation and is not load-bearing in the way of an imported uniqueness theorem. However, one part of the paper's interpretation is self-definitional: MORDA's ability to 'preview' South Korea is built into the dataset by construction, because the digital-twin maps were created from South Korean HD map, survey, and 3D-building data. The paper also calls AI-Hub 'unforeseen' while the target geography was used to design the training data. The causal claim that the Korean digital-twin content, rather than generic synthetic-data augmentation, drives the AI-Hub gains is not tested: Table IV compares other synthetic datasets only on nuScenes, and Section VI explicitly defers isolating MORDA's contributing characteristics. This is a missing-control and attribution gap plus partial by-construction circularity, but it does not reduce the central empirical improvement to a tautology.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the fidelity of the MORAI simulator's digital-twin maps and sensor/label replication, and on the comparability of the processed AI-Hub labels. These are domain assumptions rather than mathematically derivable facts; they are stated in the paper but not independently verified.

free parameters (2)
  • Static-to-dynamic scene ratio = 57 static : 30 dynamic
    Hand-chosen split of scene types in MORDA, stated in Section III-D. No ablation justifies this ratio, and it could materially affect object class and density distributions in the training mix.
  • Traffic density level = LOS E or F
    The traffic generator was set to near-capacity or breakdown flow, a manually chosen operating point with no ablation showing it is optimal for transfer, as described in Section III-D.
assumptions (4)
  • domain assumption Digital-twin maps built from HD map, survey, and 3D building data faithfully replicate real South Korean driving environments relevant to object detection.
    Section III-C states the maps were constructed from these data; the transfer results depend on this fidelity.
  • domain assumption MORAI SIM faithfully reproduces the nuScenes sensor suite (six 1600x900 cameras, 32-channel LiDAR) and 3D BBox labeling rules, including articulated vehicles.
    Section III-B; if sensor or label replication is imperfect, compatibility between MORDA and nuScenes, which underpins the training mix, is weakened.
  • domain assumption Pseudo-2D bounding boxes generated from AI-Hub panoptic segmentation are accurate enough to evaluate 2D detectors.
    Section V-B.2 states pseudo-2D labels were generated because AI-Hub lacks 2D boxes; noisy labels could bias absolute AP values, though both baseline and MORDA-trained models are affected.
  • domain assumption The four AI-Hub classes (car, truck, bus, pedestrian) have annotation policies compatible with nuScenes after preprocessing.
    Section V-B.2 says classes with deviating policies or few instances were excluded; the remaining set is assumed comparable.

how reviews work

0 comments
Cite this review

Pith. "Pith review of MORDA: A Synthetic Dataset to Facilitate Adaptation of Object Detectors to Unseen Real-target Domain While Preserving Performance on Real-source Domain." pith.science (2026). https://pith.science/paper/4SLGF3ET

@misc{pith2026250104950,
  author       = {Pith},
  title        = {Pith review of: MORDA: A Synthetic Dataset to Facilitate Adaptation of Object Detectors to Unseen Real-target Domain While Preserving Performance on Real-source Domain},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4SLGF3ET}},
  note         = {Machine review of arXiv:2501.04950}
}
read the original abstract

Deep neural network (DNN) based perception models are indispensable in the development of autonomous vehicles (AVs). However, their reliance on large-scale, high-quality data is broadly recognized as a burdensome necessity due to the substantial cost of data acquisition and labeling. Further, the issue is not a one-time concern, as AVs might need a new dataset if they are to be deployed to another region (real-target domain) that the in-hand dataset within the real-source domain cannot incorporate. To mitigate this burden, we propose leveraging synthetic environments as an auxiliary domain where the characteristics of real domains are reproduced. This approach could enable indirect experience about the real-target domain in a time- and cost-effective manner. As a practical demonstration of our methodology, nuScenes and South Korea are employed to represent real-source and real-target domains, respectively. That means we construct digital twins for several regions of South Korea, and the data-acquisition framework of nuScenes is reproduced. Blending the aforementioned components within a simulator allows us to obtain a synthetic-fusion domain in which we forge our novel driving dataset, MORDA: Mixture Of Real-domain characteristics for synthetic-data-assisted Domain Adaptation. To verify the value of synthetic features that MORDA provides in learning about driving environments of South Korea, 2D/3D detectors are trained solely on a combination of nuScenes and MORDA. Afterward, their performance is evaluated on the unforeseen real-world dataset (AI-Hub) collected in South Korea. Our experiments present that MORDA can significantly improve mean Average Precision (mAP) on AI-Hub dataset while that on nuScenes is retained or slightly enhanced.

Figures

Figures reproduced from arXiv: 2501.04950 by the authors.

Figure 1
Figure 1. Research overview: Cost-effective generation of virtual data S [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Architecture of the proposed method to generate MORDA from synthetic-fusion domain ( [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Distribution of 3D-BBox annotations by category for nuScenes [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Scatter plot of 2D-BBox labels for front-camera images in nuScenes [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 6
Figure 6. Figure 6: Enhanced generalization performance of CenterPoint-Voxel (0.1m) [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

34 extracted references · 26 canonical work pages

  1. [1]

    FCOS: Fully convolutional one- stage object detection,

    Z. Tian, C. Shen, H. Chen, and T. He, “FCOS: Fully convolutional one- stage object detection,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , 2019

    work page 2019

  2. [2]

    SMOKE: Single-Stage Monocular 3D Object Detection via Keypoint Estimation,

    Z. Liu, Z. Wu, and R. Tóth, “SMOKE: Single-Stage Monocular 3D Object Detection via Keypoint Estimation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2020, pp. 996–997

    work page 2020

  3. [3]

    DETR3D: 3D Object Detection from Multi-view Images via 3D-to- 2D Queries,

    Y . Wang, V . Guizilini, T. Zhang, Y . Wang, H. Zhao, and J. M. Solomon, “DETR3D: 3D Object Detection from Multi-view Images via 3D-to- 2D Queries,” in The Conference on Robot Learning (CoRL) , 2021

    work page 2021

  4. [4]

    PointPillars: Fast Encoders for Object Detection from Point Clouds,

    A. H. Lang, S. V ora, H. Caesar, L. Zhou, J. Yang, and O. Beijbom, “PointPillars: Fast Encoders for Object Detection from Point Clouds,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2019, pp. 12 697–12 705

    work page 2019

  5. [5]

    SSN: Shape Signature Networks for Multi-class Object Detection from Point Clouds,

    X. Zhu, Y . Ma, T. Wang, Y . Xu, J. Shi, and D. Lin, “SSN: Shape Signature Networks for Multi-class Object Detection from Point Clouds,” in Proceedings of the European Conference on Computer Vision, 2020

    work page 2020

  6. [6]

    Center-based 3d object detection and tracking,

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

    work page 2021

  7. [7]

    RadarDistill: Boosting Radar-based Object Detection Performance via Knowledge Distillation from LiDAR Features

    G. Bang, K. Choi, J. Kim, D. Kum, and J. W. Choi, “RadarDistill: Boosting Radar-based Object Detection Performance via Knowledge Distillation from LiDAR Features,” 2024. [Online]. Available: https://arxiv.org/abs/2403.05061

    work page Pith review arXiv 2024

  8. [8]

    CRN: Camera Radar Net for Accurate, Robust, Efficient 3D Perception,

    Y . Kim, J. Shin, S. Kim, I.-J. Lee, J. W. Choi, and D. Kum, “CRN: Camera Radar Net for Accurate, Robust, Efficient 3D Perception,” in Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 17 615–17 626

    work page 2023

Show all 34 references
  1. [9]

    BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird’s-Eye View Representation,

    Z. Liu, H. Tang, A. Amini, X. Yang, H. Mao, D. Rus, and 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

  2. [10]

    Cross Modal Transformer via Coordinates Encoding for 3D Object Dectection,

    J. Yan, Y . Liu, J. Sun, F. Jia, S. Li, T. Wang, and X. Zhang, “Cross Modal Transformer via Coordinates Encoding for 3D Object Dectection,” arXiv preprint arXiv:2301.01283 , 2023

    2023 arXiv

  3. [11]

    ST3D: Self-training for Unsupervised Domain Adaptation on 3D Object Detection,

    J. Yang, S. Shi, Z. Wang, H. Li, and X. Qi, “ST3D: Self-training for Unsupervised Domain Adaptation on 3D Object Detection,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021

    2021

  4. [12]

    ST3D++: Denoised Self-training for Unsupervised Domain Adaptation on 3D Object Detection,

    Jihan Yang, S. Shi, Z. Wang, H. Li, and X. Qi, “ST3D++: Denoised Self-training for Unsupervised Domain Adaptation on 3D Object Detection,” 2021. [Online]. Available: https://arxiv.org/abs/2108.06682

    2021 arXiv

  5. [13]

    Playing for benchmarks,

    S. R. Richter, Z. Hayder, and V . Koltun, “Playing for benchmarks,” in IEEE International Conference on Computer Vision, ICCV 2017, V enice, Italy, October 22-29, 2017 , 2017, pp. 2232–2241. [Online]. Available: https://doi.org/10.1109/ICCV .2017.243

    2017 doi

  6. [14]

    Virtual KITTI 2,

    Y . Cabon, N. Murray, and M. Humenberger, “Virtual KITTI 2,” 2020

    2020

  7. [15]

    Are we ready for autonomous driving? The KITTI vision benchmark suite,

    A. Geiger, P. Lenz, and R. Urtasun, “Are we ready for autonomous driving? The KITTI vision benchmark suite,” in 2012 IEEE Confer- ence on Computer Vision and Pattern Recognition , 2012, pp. 3354– 3361

    2012

  8. [16]

    Temporal coherence for active learning in videos,

    J. Zolfaghari Bengar, A. Gonzalez-Garcia, G. Villalonga, B. Raducanu, H. Habibi Aghdam, M. Mozerov, A. M. Lopez, and J. van de Weijer, “Temporal coherence for active learning in videos,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshop...

    2019

  9. [17]

    SHIFT: A Synthetic Driving Dataset for Continuous Multi-Task Domain Adaptation,

    T. Sun, M. Segu, J. Postels, Y . Wang, L. Van Gool, B. Schiele, F. Tombari, and F. Yu, “SHIFT: A Synthetic Driving Dataset for Continuous Multi-Task Domain Adaptation,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2022, pp. ...

    2022

  10. [18]

    OPV2V: An Open Benchmark Dataset and Fusion Pipeline for Perception with Vehicle-to-Vehicle Communication,

    R. Xu, H. Xiang, X. Xia, X. Han, J. Li, and J. Ma, “OPV2V: An Open Benchmark Dataset and Fusion Pipeline for Perception with Vehicle-to-Vehicle Communication,” in 2022 International Conference on Robotics and Automation (ICRA) . IEEE Press, 2022, p. 2583–2589. [Online]. Availa...

    2022

  11. [19]

    nuScenes: A multimodal dataset for autonomous driving,

    H. Caesar, V . Bankiti, A. H. Lang, S. V ora, V . E. Liong, Q. Xu, A. Krishnan, Y . Pan, G. Baldan, and O. Beijbom, “nuScenes: A multimodal dataset for autonomous driving,” in CVPR, 2020

    2020

  12. [20]

    Playing for Data: Ground Truth from Computer Games,

    S. R. Richter, V . Vineet, S. Roth, and V . Koltun, “Playing for Data: Ground Truth from Computer Games,” in European Conference on Computer Vision (ECCV) , ser. LNCS, B. Leibe, J. Matas, N. Sebe, and M. Welling, Eds., vol. 9906. Springer International Publishing, 2016, pp. 102–118

    2016

  13. [21]

    Virtual worlds as proxy for multi-object tracking analysis,

    A. Gaidon, Q. Wang, Y . Cabon, and E. Vig, “Virtual worlds as proxy for multi-object tracking analysis,” in Proceedings of the IEEE conference on Computer Vision and Pattern Recognition , 2016, pp. 4340–4349

    2016

  14. [22]

    A theory of learning from different domains,

    S. Ben-David, J. Blitzer, K. Crammer, A. Kulesza, F. C. Pereira, and J. W. Vaughan, “A theory of learning from different domains,” Machine Learning , vol. 79, pp. 151–175, 2010. [Online]. Available: https://api.semanticscholar.org/CorpusID:8577357

    2010

  15. [23]

    UADA3D: Unsupervised Adversarial Domain Adaptation for 3D Object Detection with Sparse LiDAR and Large Domain Gaps,

    M. K. Wozniak, M. Hansson, M. Thiel, and P. Jensfelt, “UADA3D: Unsupervised Adversarial Domain Adaptation for 3D Object Detection with Sparse LiDAR and Large Domain Gaps,” 2024. [Online]. Available: https://arxiv.org/abs/2403.17633

    2024 arXiv

  16. [24]

    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

  17. [25]

    ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentation,

    T.-H. Vu, H. Jain, M. Bucher, M. Cord, and P. Pérez, “ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentation,” in CVPR, 2019

    2019

  18. [26]

    On Learning Invariant Representations for Domain Adaptation,

    H. Zhao, R. T. des Combes, K. Zhang, and G. J. Gordon, “On Learning Invariant Representations for Domain Adaptation,” in International Conference on Machine Learning , 2019. [Online]. Available: https://api.semanticscholar.org/CorpusID:174800477

    2019

  19. [27]

    Unsupervised Intra-domain Adaptation for Semantic Segmentation through Self- Supervision,

    F. Pan, I. Shin, F. Rameau, S. Lee, and I. S. Kweon, “Unsupervised Intra-domain Adaptation for Semantic Segmentation through Self- Supervision,” in IEEE Conference on Computer Vision and Pattern Recoginition (CVPR) , 2020

    2020

  20. [28]

    FDA: Fourier Domain Adaptation for Semantic Segmentation,

    Y . Yang and S. Soatto, “FDA: Fourier Domain Adaptation for Semantic Segmentation,” 2020. [Online]. Available: https: //arxiv.org/abs/2004.05498

    2020 arXiv

  21. [29]

    CARLA: An open urban driving simulator,

    A. Dosovitskiy, G. Ros, F. Codevilla, A. Lopez, and V . Koltun, “CARLA: An open urban driving simulator,” in Proceedings of the 1st Annual Conference on Robot Learning , ser. Proceedings of Machine Learning Research, S. Levine, V . Vanhoucke, and K. Goldberg, Eds., vol. 78. PM...

    2017

  22. [30]

    MORAI Simulator,

    “MORAI Simulator,” https://www.morai.ai/, MORAI Inc

  23. [31]

    Ministry of land,

    K. H. C. Manual, “Ministry of land,” Transport and Maritime Affairs , pp. p210–417, 2013

    2013

  24. [32]

    Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,

    S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” IEEE Transactions on Pattern Analysis and Machine Intelligence , Jun 2017

    2017

  25. [33]

    MMDetection: Open mmlab detection toolbox and benchmark,

    K. Chen, J. Wang, J. Pang, Y . Cao, Y . Xiong, X. Li, S. Sun, W. Feng, Z. Liu, J. Xu, Z. Zhang, D. Cheng, C. Zhu, T. Cheng, Q. Zhao, B. Li, X. Lu, R. Zhu, Y . Wu, J. Dai, J. Wang, J. Shi, W. Ouyang, C. C. Loy, and D. Lin, “MMDetection: Open mmlab detection toolbox and benchmar...

    1906 arXiv

  26. [34]

    MMDetection3D: OpenMMLab next-generation platform for general 3D object detection,

    M. Contributors, “MMDetection3D: OpenMMLab next-generation platform for general 3D object detection,” https://github.com/ open-mmlab/mmdetection3d, 2020

    2020

Pith tools

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