IROAM: Improving Roadside Monocular 3D Object Detection Learning from Autonomous Vehicle Data Domain

Jingjing Liu; Siqi Fan; Xiaoliang Huo; Yan Wang; Ya-Qin Zhang; Zhe Wang

REVIEW 4 major objections 5 minor 27 references

IROAM: Improving Roadside Monocular 3D Object Detection Learning from Autonomous Vehicle Data Domain

T0 review · 4 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 IROAM improves roadside monocular 3D detection by aligning only semantic object queries across vehicle-side and roadside views, lifting AP3D at IoU 0.7 from 24.31 to 28.89 on the easy split.

desk verdict A clean but narrow contrastive-learning trick for roadside 3D detection; the strong 'abundant AV data' claim is not yet backed by the experiments. read the letter →

arxiv 2501.18162 v1 pith:SPDXJLTE submitted 2025-01-30 cs.CV cs.RO

Zhe Wang , Xiaoliang Huo , Siqi Fan , Jingjing Liu , Ya-Qin Zhang , Yan Wang This is my paper · ORCID

classification cs.CVcs.RO

keywords roadsidemonocular3Dobjectdetectioncontrastivelearningsemantic-geometrydecouplingcross-domainvehicle-infrastructurecooperationqueriesdepth-guidedtransformer

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 paper argues that roadside monocular 3D object detectors can be made more accurate by training them jointly on abundant vehicle-mounted camera data, despite a substantial viewpoint gap between the two domains. The central idea is to extract object queries from both domains and align only their semantic parts across views, leaving geometric parts such as depth and orientation to be supervised separately within each domain. On the DAIR-V2X vehicle-infrastructure benchmark, the proposed IROAM method raises roadside AP3D at IoU 0.7 from 24.31 to 28.89 on the easy split, and from 13.54 to 16.11 on the moderate split. The gains hold when the vehicle-side dataset is swapped, suggesting cross-domain transfer rather than overfitting.

What carries the argument

The load-bearing mechanism is semantic-geometry decoupled contrastive learning over transformer object queries. Each query produced by a depth-guided detector is split into semantic and geometry halves; cosine similarity between semantic halves of positive cross-domain pairs is pulled toward 1, negative pairs toward 0, while geometry halves are never compared across domains. A Query Sampler selects positive and negative samples using Hungarian matching scores in each domain. This lets the model borrow category and appearance knowledge from the vehicle domain without forcing depth and geometry distributions to match.

What would settle it

Rerun IROAM's training on the same benchmark but with the vehicle-side and roadside frames randomly re-paired within each batch; if the roadside AP3D improvement over the data-merging baseline remains, then the claimed benefit does not depend on real cross-view correspondences, contradicting the method's design.

Watch

Extended reading notes

Core claim

The paper's discovery is that the reason vehicle-side data has not helped roadside detectors is not a lack of useful information but a geometry mismatch: the same object appears with very different depth distributions from the two viewpoints. IROAM therefore decouples each object query into a semantic half and a geometry half, runs contrastive learning only on the semantic half, and keeps geometry supervised per domain. This yields roadside AP3D(IoU=0.7) of 28.89/16.11/15.85 on easy/moderate/hard, compared with 24.31/13.54/13.17 for the roadside-only baseline and 26.74/15.36/13.84 for naive data merging. The same training also improves vehicle-side detection, and the benefit survives replacing the vehicle-side dataset, supporting the claim that semantic alignment transfers across vehicle and roadside domains.

Load-bearing premise

The method needs paired, synchronized vehicle-side and roadside images with ground-truth depth in both domains during training, so it does not apply as described to the common case of unpaired or depth-unlabeled vehicle-side data.

Editorial extensions

If this is right

Roadside perception for vehicle-infrastructure cooperative driving can be improved by reusing existing vehicle-side datasets, reducing the need for costly roadside annotation.
Cross-domain training also improves vehicle-side detection, so the benefit is bidirectional rather than one-way transfer.
The improvement persists when the vehicle-side dataset is changed, indicating the learned semantic alignment generalizes across datasets.
The method remains useful when roadside data is scarce relative to vehicle-side data, with no extra inference cost.
Decoupling semantic from geometry contributes roughly 1.99 AP3D over contrastive learning on full queries, isolating the mechanism that drives the gain.

Reading between the lines

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

If semantic-geometry decoupling is the active ingredient, the same recipe may transfer to other viewpoint gaps, such as drone-to-ground cameras or different camera heights, where semantics match but geometry does not.
Because the contrastive loss operates on object queries rather than pixels, it could compose with any query-based detector; a natural test is swapping the depth-guided backbone for a lighter query-based detector.
A testable extension is to relax the paired-synchronization requirement using pseudo-pairs or self-supervised depth, which would let the method exploit large unpaired vehicle-side collections.
The imbalance experiment suggests a fixed pool of vehicle-side data can serve many roadside locations, pointing toward pretraining a roadside detector once on a large vehicle-side corpus.

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

Read it. The method is a sensible extension of the MonoDETR line, and the reported gains on roadside monocular 3D detection are consistent. But be careful with the framing: the paper sells itself as exploiting abundant vehicle-side data while the experiments only use vehicle-side data from vehicle-infrastructure cooperative datasets. That is a weaker claim than advertised.

The new trick is to split each DETR query into a semantic half and a geometry half, then run contrastive learning only on the semantic half. That is a small but clean idea, and the ablation shows it matters: adding the decoupling raises AP3D(IoU=0.7) moderate from 15.73 to 16.11, and the full method beats both roadside-only and the naive Addon baseline on every metric. The imbalance analysis (Fig. 5) is also a nice touch; it shows IROAM is more compute-efficient than Addon when vehicle-side data dominates.

Three weaknesses. First, no experiment on generic AV data. KITTI, nuScenes, and Waymo are all named as motivation, but all training uses paired VIC vehicle-side data with depth supervision. The method does not need temporally synchronized vehicle/roadside frames—it just needs one road and one vehicle image per batch, and depth labels are available in KITTI/nuScenes/Waymo from LiDAR. So the stress-test note overstates the pairing requirement, but the underlying point stands: the open question is whether IROAM transfers to truly diverse, non-VIC vehicle data. An experiment with KITTI as vehicle side would settle it.

Second, the contrastive objective in Eq. 5 labels every pair of positive queries as a positive, regardless of whether the objects are the same category. The model pulls all objects in an image toward one another, which seems likely to wash out object-class distinctions. The L1 loss on a sigmoid cosine similarity is also under-justified. An ablation with a class-aware contrastive loss, or at least a discussion, is needed.

Third, all numbers are single-run on the val split. Gains of 4-5 AP are large, but no error bars means I can't tell if the edge over Addon is meaningful. The paper should rerun with a few seeds. Also, no code release, which makes these issues harder to check.

Bottom line: I'd send it to review. The core idea is publishable and the experiments, though narrow, are internally coherent. The authors need to broaden the evaluation, tighten the contrastive loss, and report variance.

Referee Report

4 major / 5 minor

Summary. The paper proposes IROAM, a framework for roadside monocular 3D object detection that jointly trains a roadside branch and a vehicle-side branch on top of MonoDETR. It introduces an In-Domain Query Interaction module that produces object queries from content and depth features, and a Cross-Domain Query Enhancement module that decouples queries into semantic and geometry parts and applies contrastive learning only to the semantic parts. Experiments on DAIR-V2X and V2X-Seq report consistent improvements over roadside-only and simple data-merging baselines on AP3D and APBEV at IoU thresholds of 0.7 and 0.5, including cross-dataset vehicle-side training and an analysis of data imbalance.

Significance. If the reported results are reproducible, the paper demonstrates a useful technique for transferring vehicle-side monocular representations to roadside detection, and the modular design plus cross-dataset experiments (DAIR-V2X and V2X-Seq) are strengths. The analysis of data imbalance in Figure 5 is a valuable addition. However, the central claim about learning from abundant vehicle-side data is not fully supported because the method as described requires paired, depth-annotated vehicle-roadside images; the claimed applicability to vehicle-only datasets such as KITTI, nuScenes, and Waymo is therefore unverified. The paper would be strengthened by a precise pairing protocol, a more standard contrastive formulation, and either experiments on unpaired data or a more careful statement of scope.

major comments (4)

[III-D] The overall loss in Eq. (7) requires, for every training step, a synchronized vehicle-roadside image pair with ground-truth depth maps and object labels in both domains. However, the Introduction and Conclusion motivate the method by "tremendous and diverse vehicle-side data" from vehicle-only datasets such as KITTI, nuScenes, and Waymo. These datasets do not contain synchronized roadside images or roadside ground truth, so IROAM cannot be trained on them as described. All experiments in Table I use DAIR-V2X-V or DAIR-Seq-V as vehicle-side data, both of which are paired with the evaluation roadside dataset. The paper does not test an alternative regime, such as pretraining the vehicle branch on unpaired vehicle data and fine-tuning on sparse vehicle-roadside pairs. The central claim that IROAM learns from the abundant autonomous-vehicle data domain is therefore not supported by the reported experiments.
[III-C] The paper does not specify how the vehicle-side and roadside images are paired. This matters because the contrastive loss in Eq. (6) treats every query matched to any ground-truth object in the merged positive set QP as mutually similar (Eq. (5), label 1 for all i,j in P). If images are randomly paired or come from different datasets (as in Table I rows that use DAIR-Seq-V with DAIR-V2X-I), then positive pairs do not correspond to the same objects or even the same scenes, and the semantic-alignment rationale illustrated in Figure 1 is not realized. Please specify the pairing procedure, whether synchronized frames from the VIC dataset are used, and whether positive labels are based on object correspondence or merely on 'any matched object'.
[III-C] In Eqs. (4)-(6), the similarity is defined as a sigmoid of the cosine similarity, so the maximum possible value of s_i,j is about 0.731 while the target label is 1 in Eq. (5). The L1 loss in Eq. (6) therefore never reaches zero for positive pairs, and the gradient behavior of this formulation is not analyzed. Additionally, the positive set QP merges queries from both domains without considering object category or identity, which may encourage alignment of different object types. Please justify the sigmoid-plus-L1 formulation or replace it with a standard InfoNCE or cosine-embedding loss, and discuss the effect of cross-category positives.
[IV-C] In Table III, the row labeled with only "Veh Data" (CL and DC absent) reports AP3D easy of 24.61, whereas the Addon baseline in Table I, which also trains on vehicle-side and roadside data together, reports 26.74. The text states that "Veh Data means to train the model with both roadside and vehicle data," which should be equivalent to Addon. The discrepancy is unexplained and makes it difficult to attribute the gains in the ablation to the individual components. Please clarify the experimental setup of Table III (e.g., number of vehicle images per epoch, use of pair sampling, and any differences from the Addon configuration in Table I) and reconcile the two tables.

minor comments (5)

[III-C] Eq. (6) contains a typo: "lable" should be "label".
[IV-A] All experiments are reported as single runs on the validation split; please provide multiple seeds or at least a statement of variance, since the reported improvements are moderate in some cells.
[Figure 2] The caption of Figure 2 is dense and does not introduce the notation f^e_{C,*} and f^e_{D,*}; please define these in the caption or point explicitly to Section III-A.
[III-B] Equations (2) and (3) use overlapping weight notation: Eq. (2) uses lambda_1..4 for matching cost while Eq. (3) adds lambda_5..7 for the final pair loss. Consider clarifying that these are two distinct weighted sums and that the same lambda symbols appear in both.
[I] The first sentence of the Introduction contains a capitalization error: "In autonomous driving, The perception..." should be "the perception...".

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: IROAM's gains are measured on held-out data with objectives independent of the evaluation metric.

full rationale

IROAM's central claim is empirical: adding semantic-only cross-domain contrastive learning to a MonoDETR-style detector improves roadside AP3D/APBEV on DAIR-V2X-I and DAIR-Seq-I validation splits. The training objective in Eq. (7) is a sum of per-domain pair losses, depth-map Focal losses, and the contrastive loss Lcl; none of these terms is defined in terms of the evaluation metric (AP under IoU thresholds), and the reported numbers are computed on held-out validation annotations with ground-truth boxes. The Query Sampler selects positive/negative sets from Hungarian matching scores, but that is an auxiliary training-time selection; it does not make the evaluation score a function of the loss. The only cited prior works that are load-bearing (MonoDETR [15] as backbone, CO^3 [25] as motivation) are externally published and not authored by the IROAM authors; no uniqueness theorem or ansatz is imported from the authors' own prior work. The paper's stated limitation—training requires paired synchronized vehicle/roadside images with depth labels (Eq. 7, Section IV-A)—is a data-regime mismatch with the 'abundant vehicle-side data' motivation, but a mismatch between motivation and method is not a circular derivation. No step reduces a prediction to a fitted input by construction, so no significant circularity is found.

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

The paper's central claim rests on several hand-chosen hyperparameters and three unverified structural assumptions: the clean semantic/geometry split of query channels, the all-positive-label contrastive target, and the availability of paired depth-supervised data. No new physical or model entities are introduced.

free parameters (5)

Loss weights λ1..λ7 = 2, 10, 5, 2, 1, 1, 1
Set by hand in Section IV-A to balance classification, 3D center, 2D edge, GIoU, dimension, orientation, and depth losses; the central detection loss depends on them.
Number of object queries N = 50
Chosen in Section IV-A; capacity of the detector and of the contrastive sample set.
Query channel dimension = 256
Chosen in Section IV-A; the split into semantic and geometric halves assumes the semantic information fits in 128 channels.
Depth supervision range = [2, 65] meters
Samples with depth outside this range are ignored during training (Section IV-A); affects the depth map and regression supervision.
Confidence threshold at inference = 0.2
Set in Section IV-A to filter object queries at inference; affects AP.

assumptions (5)

ad hoc to paper The first N/2 channels of each query encode semantic content and the latter N/2 encode geometry.
Section III-C assumes a hard split of the 256-dim query into semantic and geometric halves with no analysis showing the split actually separates these properties; if the assumption fails, the contrastive objective is misaligned.
ad hoc to paper All positive queries from both domains are mutually similar, regardless of object identity or category.
Eq. (5) labels every positive-positive pair as 1, which forces semantic features of different objects toward each other; the paper does not justify this beyond asserting semantic generality, and it is not a standard contrastive formulation.
domain assumption Paired, synchronized vehicle and roadside images with depth ground truth are available for both domains.
Eq. (7) sums losses for a pair of roadside and vehicle images; depth maps are supervised by LiDAR-derived ground truth in both branches (Section III-A). The Introduction motivates using vehicle-side data, but the method assumes labels and pairing, which may not scale.
ad hoc to paper Hungarian matching scores select meaningful positive and negative query sets.
The Query Sampler in Section III-C takes the Kr highest-scoring matched queries as positives and the Kr lowest as negatives; there is no validation that these correlate with useful semantic features.
standard math Ground-truth depth maps can be generated from LiDAR for both datasets.
DAIR-V2X and V2X-Seq provide LiDAR; the paper relies on foreground depth maps D_fg, supervised by LiDAR-derived depth (Section III-A). This is a standard assumption for these datasets.

how reviews work

0 comments

Cite this review

Pith. "Pith review of IROAM: Improving Roadside Monocular 3D Object Detection Learning from Autonomous Vehicle Data Domain." pith.science (2026). https://pith.science/paper/SPDXJLTE

@misc{pith2026250118162,
  author       = {Pith},
  title        = {Pith review of: IROAM: Improving Roadside Monocular 3D Object Detection Learning from Autonomous Vehicle Data Domain},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SPDXJLTE}},
  note         = {Machine review of arXiv:2501.18162}
}

read the original abstract

In autonomous driving, The perception capabilities of the ego-vehicle can be improved with roadside sensors, which can provide a holistic view of the environment. However, existing monocular detection methods designed for vehicle cameras are not suitable for roadside cameras due to viewpoint domain gaps. To bridge this gap and Improve ROAdside Monocular 3D object detection, we propose IROAM, a semantic-geometry decoupled contrastive learning framework, which takes vehicle-side and roadside data as input simultaneously. IROAM has two significant modules. In-Domain Query Interaction module utilizes a transformer to learn content and depth information for each domain and outputs object queries. Cross-Domain Query Enhancement To learn better feature representations from two domains, Cross-Domain Query Enhancement decouples queries into semantic and geometry parts and only the former is used for contrastive learning. Experiments demonstrate the effectiveness of IROAM in improving roadside detector's performance. The results validate that IROAM has the capabilities to learn cross-domain information.

Figures

Figures reproduced from arXiv: 2501.18162 by the authors.

**Figure 2.** The framework of IROAM contains a roadside branch and a vehicle-side branch and each branch has the same architecture of Feature Encoder [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** The procedure of semantic-geometry decoupled Contrastive [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Visualization results of IROAM. From BEV, it is clear that prediction bounding boxes (red) and labels (green) from IROAM are better aligned [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]

**Figure 5.** Figure 5: Analysis on different proportion ratios of vehicle-side to roadside [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 14 canonical work pages

[1]

Dair-v2x: A large-scale dataset for vehicle- infrastructure cooperative 3d object detection,

H. Yu, Y . Luo, M. Shu, Y . Huo, Z. Yang, Y . Shi, Z. Guo, H. Li, X. Hu, J. Yuan, and Z. Nie, “Dair-v2x: A large-scale dataset for vehicle- infrastructure cooperative 3d object detection,” in IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR) , June 2022

work page 2022
[2]

V2x-seq: A large- scale sequential dataset for vehicle-infrastructure cooperative percep- tion and forecasting,

H. Yu, W. Yang, H. Ruan, Z. Yang, Y . Tang, X. Gao, X. Hao, Y . Shi, Y . Pan, N. Sun, J. Song, J. Yuan, P. Luo, and Z. Nie, “V2x-seq: A large- scale sequential dataset for vehicle-infrastructure cooperative percep- tion and forecasting,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2023

2023
[3]

Rope3d: The roadside perception dataset for autonomous driving and monocular 3d object detection task,

X. Ye, M. Shu, H. Li, Y . Shi, Y . Li, G. Wang, X. Tan, and E. Ding, “Rope3d: The roadside perception dataset for autonomous driving and monocular 3d object detection task,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2022, pp. 21 341–21 350

2022
[4]

Bevheight: A robust framework for vision-based roadside 3d object detection,

L. Yang, K. Yu, T. Tang, J. Li, K. Yuan, L. Wang, X. Zhang, and P. Chen, “Bevheight: A robust framework for vision-based roadside 3d object detection,” in IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR), Mar. 2023

work page 2023
[5]

Vision meets robotics: The kitti dataset,

A. Geiger, P. Lenz, C. Stiller, and R. Urtasun, “Vision meets robotics: The kitti dataset,” International Journal of Robotics Research (IJRR) , 2013

2013
[6]

nuscenes: A multimodal dataset for autonomous driving,

H. Caesar, V . Bankiti, A. H. Lang, S. V ora, V . E. Liong, Q. Xu, A. Kr- ishnan, Y . Pan, G. Baldan, and O. Beijbom, “nuscenes: A multimodal dataset for autonomous driving,” arXiv preprint arXiv:1903.11027 , 2019

arXiv 1903
[7]

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 perception for autonomous driving: Waymo open dataset,” in Proceed- ings of the IEEE...

work page 2020
[8]

Faster R-CNN: towards real-time object detection with region proposal networks,

S. Ren, K. He, R. B. Girshick, and J. Sun, “Faster R-CNN: towards real-time object detection with region proposal networks,” IEEE Trans. Pattern Anal. Mach. Intell. , vol. 39, no. 6, pp. 1137–1149, 2017. [Online]. Available: https://doi.org/10.1109/TPAMI.2016.2577031

arXiv 2017

Show all 27 references

[9]

Focal loss for dense object detection,

T. Lin, P. Goyal, R. B. Girshick, K. He, and P. Doll ´ar, “Focal loss for dense object detection,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 42, no. 2, pp. 318–327, 2020. [Online]. Available: https://doi.org/10.1109/TPAMI.2018.2858826

2020
[10]

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 , 2019, pp. 9627–9636

2019
[11]

Objects as points,

X. Zhou, D. Wang, and P. Kr ¨ahenb¨uhl, “Objects as points,” arXiv preprint arXiv:1904.07850, 2019

1904 arXiv
[12]

Attention is all you need,

A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” Advances in neural information processing systems , vol. 30, 2017

2017
[13]

End-to-end object detection with transformers,

N. Carion, F. Massa, G. Synnaeve, N. Usunier, A. Kirillov, and S. Zagoruyko, “End-to-end object detection with transformers,” in European conference on computer vision . Springer, 2020, pp. 213– 229

2020
[14]

Deformable DETR: deformable transformers for end-to-end object detection,

X. Zhu, W. Su, L. Lu, B. Li, X. Wang, and J. Dai, “Deformable DETR: deformable transformers for end-to-end object detection,” in 9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021 . OpenReview.net, 2021. [Online]. Availab...

2021
[15]

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

R. Zhang, H. Qiu, T. Wang, X. Xu, Z. Guo, Y . Qiao, P. Gao, and H. Li, “Monodetr: Depth-guided transformer for monocular 3d object detection,” ICCV 2023, 2022

2023
[16]

A simple framework for contrastive learning of visual representations,

T. Chen, S. Kornblith, M. Norouzi, and G. E. Hinton, “A simple framework for contrastive learning of visual representations,” in Proceedings of the 37th International Conference on Machine Learning, ICML 2020, 13-18 July 2020, Virtual Event , ser. Proceedings of Machine Learni...

2020
[17]

Aligning pretraining for detection via object-level contrastive learning,

F. Wei, Y . Gao, Z. Wu, H. Hu, and S. Lin, “Aligning pretraining for detection via object-level contrastive learning,” in Advances in Neural Information Processing Systems 34: Annual Conference on Neural Information Processing Systems 2021, NeurIPS 2021, December 6-14, 2021, v...

2021
[18]

Simipu: Simple 2d image and 3d point cloud unsupervised pre-training for spatial-aware visual representations,

Z. Li, Z. Chen, A. Li, L. Fang, Q. Jiang, X. Liu, J. Jiang, B. Zhou, and H. Zhao, “Simipu: Simple 2d image and 3d point cloud unsupervised pre-training for spatial-aware visual representations,” in Thirty-Sixth AAAI Conference on Artificial Intelligence, AAAI 2022, Thirty-Four...

2022
[19]

4dcontrast: Contrastive learning with dynamic correspondences for 3d scene understanding,

Y . Chen, M. Nießner, and A. Dai, “4dcontrast: Contrastive learning with dynamic correspondences for 3d scene understanding,” in Computer Vision - ECCV 2022 - 17th European Conference, Tel Aviv, Israel, October 23-27, 2022, Proceedings, Part XXXII , ser. Lecture Notes in Compu...

2022 doi
[20]

Momentum contrast for unsupervised visual representation learning,

K. He, H. Fan, Y . Wu, S. Xie, and R. B. Girshick, “Momentum contrast for unsupervised visual representation learning,” in 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2020, Seattle, WA, USA, June 13-19, 2020 . Computer Vision Foundation / IEEE, 20...

2020
[21]

Contrastive multiview coding,

Y . Tian, D. Krishnan, and P. Isola, “Contrastive multiview coding,” in Computer Vision - ECCV 2020 - 16th European Conference, Glasgow, UK, August 23-28, 2020, Proceedings, Part XI , ser. Lecture Notes in Computer Science, A. Vedaldi, H. Bischof, T. Brox, and J. Frahm, Eds., ...

2020 doi
[22]

Unsupervised learning of visual features by contrasting cluster assignments,

M. Caron, I. Misra, J. Mairal, P. Goyal, P. Bojanowski, and A. Joulin, “Unsupervised learning of visual features by contrasting cluster assignments,” in Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurI...

2020
[23]

Bootstrap your own latent - A new approach to self-supervised learning,

J. Grill, F. Strub, F. Altch ´e, C. Tallec, P. H. Richemond, E. Buchatskaya, C. Doersch, B. ´A. Pires, Z. Guo, M. G. Azar, B. Piot, K. Kavukcuoglu, R. Munos, and M. Valko, “Bootstrap your own latent - A new approach to self-supervised learning,” in Advances in Neural Informati...

2020
[24]

Dense contrastive learning for self-supervised visual pre-training,

X. Wang, R. Zhang, C. Shen, T. Kong, and L. Li, “Dense contrastive learning for self-supervised visual pre-training,” in IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2021, virtual, June 19-25, 2021 . Computer Vision Foundation / IEEE, 2021, pp. 3024–3033. [...

2021
[25]

Coˆ 3: Cooperative unsupervised 3d representation learning for autonomous driving,

R. Chen, Y . Mu, R. Xu, W. Shao, C. Jiang, H. Xu, Z. Li, and P. Luo, “Coˆ 3: Cooperative unsupervised 3d representation learning for autonomous driving,” arXiv preprint arXiv:2206.04028 , 2022

2022
[26]

Deep residual learning for image recognition,

K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition , 2016, pp. 770–778. 7

2016
[2022]

1500–1508

AAAI Press, 2022, pp. 1500–1508. [Online]. Available: https://doi.org/10.1609/aaai.v36i2.20040

2022 doi

Pith tools

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

[1] [1]

Dair-v2x: A large-scale dataset for vehicle- infrastructure cooperative 3d object detection,

H. Yu, Y . Luo, M. Shu, Y . Huo, Z. Yang, Y . Shi, Z. Guo, H. Li, X. Hu, J. Yuan, and Z. Nie, “Dair-v2x: A large-scale dataset for vehicle- infrastructure cooperative 3d object detection,” in IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR) , June 2022

work page 2022

[2] [2]

V2x-seq: A large- scale sequential dataset for vehicle-infrastructure cooperative percep- tion and forecasting,

H. Yu, W. Yang, H. Ruan, Z. Yang, Y . Tang, X. Gao, X. Hao, Y . Shi, Y . Pan, N. Sun, J. Song, J. Yuan, P. Luo, and Z. Nie, “V2x-seq: A large- scale sequential dataset for vehicle-infrastructure cooperative percep- tion and forecasting,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2023

2023

[3] [3]

Rope3d: The roadside perception dataset for autonomous driving and monocular 3d object detection task,

X. Ye, M. Shu, H. Li, Y . Shi, Y . Li, G. Wang, X. Tan, and E. Ding, “Rope3d: The roadside perception dataset for autonomous driving and monocular 3d object detection task,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2022, pp. 21 341–21 350

2022

[4] [4]

Bevheight: A robust framework for vision-based roadside 3d object detection,

L. Yang, K. Yu, T. Tang, J. Li, K. Yuan, L. Wang, X. Zhang, and P. Chen, “Bevheight: A robust framework for vision-based roadside 3d object detection,” in IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR), Mar. 2023

work page 2023

[5] [5]

Vision meets robotics: The kitti dataset,

A. Geiger, P. Lenz, C. Stiller, and R. Urtasun, “Vision meets robotics: The kitti dataset,” International Journal of Robotics Research (IJRR) , 2013

2013

[6] [6]

nuscenes: A multimodal dataset for autonomous driving,

H. Caesar, V . Bankiti, A. H. Lang, S. V ora, V . E. Liong, Q. Xu, A. Kr- ishnan, Y . Pan, G. Baldan, and O. Beijbom, “nuscenes: A multimodal dataset for autonomous driving,” arXiv preprint arXiv:1903.11027 , 2019

arXiv 1903

[7] [7]

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 perception for autonomous driving: Waymo open dataset,” in Proceed- ings of the IEEE...

work page 2020

[8] [8]

Faster R-CNN: towards real-time object detection with region proposal networks,

S. Ren, K. He, R. B. Girshick, and J. Sun, “Faster R-CNN: towards real-time object detection with region proposal networks,” IEEE Trans. Pattern Anal. Mach. Intell. , vol. 39, no. 6, pp. 1137–1149, 2017. [Online]. Available: https://doi.org/10.1109/TPAMI.2016.2577031

arXiv 2017