Skip to content

Latest commit

 

History

History
77 lines (71 loc) · 12.8 KB

real_world_methods.md

File metadata and controls

77 lines (71 loc) · 12.8 KB
Raw

A summary of state-of-the-art collaborative perception methods for real-world issues

Table

Method Venue Modality Scheme Loc Error Comm Issue Discrep Security Code
RobustV2VNet [1] CoRL'20 LiDAR I Loc, Pos - - - -
AOMAC [2] ICCV'21 LiDAR I - - - Attack -
P-CNN [3] IoT'21 Camera E - - - Privacy -
FPV-RCNN [4] RAL'22 LiDAR I Loc, Pos - - - Linkn
TCLF [5] CVPR'22 LiDAR L - Laten - - Linkv
V2X-ViT [6] ECCV'22 LiDAR I Loc, Pos Laten - - Linko
SyncNet [7] ECCV'22 LiDAR I - Laten - - Linkc
TaskAgnostic [8] CoRL'22 LiDAR I - - Task - Linkc
SecPCV [9] TITS'22 LiDAR E - - - Privacy -
ModelAgnostic [10] ICRA'23 LiDAR L - - Model - Linko
MPDA [11] ICRA'23 LiDAR I - - Model - Linko
CoAlign [12] ICRA'23 LiDAR I, L Loc, Pos - - - Linko
LCRN [13] TIV'23 LiDAR L - Loss - - -
OptiMatch [14] IV'23 LiDAR L Loc, Pos - - - -
P2OD [15] IoT'23 Camera E - - - Privacy -
ROBOSAC [16] ICCV'23 LiDAR I - - - Attack Linkc
FFNet [17] NeurIPS'23 LiDAR I - Laten - - Linkv
CoBEVFlow [18] NeurIPS'23 LiDAR I - Laten - - Linko
How2comm [19] NeurIPS'23 LiDAR I - Laten - - Linko
FeaCo [20] MM'23 LiDAR I Loc, Pos - - - Linko
ERMVP [21] CVPR'24 LiDAR I Loc, Pos - - - Linko
MRCNet [22] CVPR'24 LiDAR I Loc, Pos - - - Linko
RoCo [23] MM'24 LiDAR I Loc, Pos - - - Linko
V2X-INCOP [a1] arXiv'23 LiDAR I - Inter - - -
MADE [a2] arXiv'23 LiDAR I - - - Attack -

Notes:

  • Schemes include early (E), intermediate (I) and late (L) collaboration.
  • Loc Error includes localization (Loc) and pose (Pos) errors.
  • Comm Issue includes latency (Laten), interruption (Inter) and loss (Loss).
  • Discrep includes model (Model) and task (Task) discrepancies.
  • Security includes attack defense (Attack) and privacy protection (Privacy).
  • Code Framework: o (OpenCOOD), v (VIC3D), c (CoPerception), n (Non-mainstream framework)

Back to Contents 🔙

References

Published

  1. Learning to Communicate and Correct Pose Errors (CoRL'20) [pdf]
  2. Adversarial attacks on multi-agent communication (ICCV'21) [pdf]
  3. Toward lightweight, privacy-preserving cooperative object classification for connected autonomous vehicles (IoT'21) [pdf]
  4. Keypoints-Based Deep Feature Fusion for Cooperative Vehicle Detection of Autonomous Driving (RAL'22) [pdf] [code]
  5. DAIR-V2X: A Large-Scale Dataset for Vehicle-Infrastructure Cooperative 3D Object Detection (CVPR'22) [pdf] [code]
  6. V2X-ViT: Vehicle-to-everything cooperative perception with vision transformer (ECCV'22) [pdf] [code]
  7. Latency-Aware Collaborative Perception (ECCV'22) [pdf] [code]
  8. Multi-robot scene completion: Towards task-agnostic collaborative perception (CoRL'22) [pdf] [code]
  9. Edge-Cooperative Privacy-Preserving Object Detection Over Random Point Cloud Shares for Connected Autonomous Vehicles (TITS'22) [pdf]
  10. Model-Agnostic Multi-Agent Perception Framework (ICRA'23) [pdf] [code]
  11. Bridging the Domain Gap for Multi-Agent Perception (ICRA'23) [pdf] [code]
  12. Robust Collaborative 3D Object Detection in Presence of Pose Errors (ICRA'23) [pdf] [code]
  13. Learning for Vehicle-to-Vehicle Cooperative Perception under Lossy Communication (TIV'23) [pdf]
  14. A Cooperative Perception System Robust to Localization Errors (IV'23) [pdf]
  15. Achieving Lightweight and Privacy-Preserving Object Detection for Connected Autonomous Vehicles (IoT'23) [pdf]
  16. Among Us: Adversarially Robust Collaborative Perception by Consensus (ICCV'23) [pdf] [code]
  17. Flow-Based Feature Fusion for Vehicle-Infrastructure Cooperative 3D Object Detection (NeurIPS'23) [pdf] [code]
  18. Robust Asynchronous Collaborative 3D Detection via Bird’s Eye View Flow (NeurIPS'23) [pdf] [code]
  19. How2comm: Communication-Efficient and Collaboration-Pragmatic Multi-Agent Perception (NeurIPS'23) [pdf] [code]
  20. FeaCo: Reaching Robust Feature-Level Consensus in Noisy Pose Conditions (MM'23) [pdf]
  21. ERMVP: Communication-Efficient and Collaboration-Robust Multi-Vehicle Perception in Challenging Environments (CVPR'24) [pdf]
  22. Multi-agent Collaborative Perception via Motion-aware Robust Communication Network (CVPR'24) [pdf]
  23. RoCo: Robust Cooperative Perception By Iterative Object Matching and Pose Adjustment (MM'24) [pdf]

ArXiv

  1. Interruption-Aware Cooperative Perception for V2X Communication-Aided Autonomous Driving (arXiv'23) [pdf]
  2. Malicious Agent Detection for Robust Multi-Agent Collaborative Perception. (arXiv'23) [pdf]