Research map / Multi-agent systems and games / Human-AI and goal-conditioned
CATNet: Collaborative Alignment and Transformation Network for Cooperative Perception
Gong Chen, Chaokun Zhang, Tao Tang, Pengcheng Lv, Feng Li, Xin Xie
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add STSync to baseline (ablation) | intermediate-fusion baseline · OPV2V | AP@0.5 / AP@0.7: 0.595/0.384 -> 0.818/0.678 | Improved |
| add STSync to baseline (ablation) | intermediate-fusion baseline · DAIR-V2X | AP@0.5 / AP@0.7: 0.659/0.461 -> 0.683/0.496 | Improved |
| add WTDen to baseline (ablation) | intermediate-fusion baseline · OPV2V | AP@0.5 / AP@0.7: 0.595/0.384 -> 0.645/0.461 | Improved |
| add AdpSel to baseline (ablation) | intermediate-fusion baseline · OPV2V | AP@0.5 / AP@0.7: 0.595/0.384 -> 0.624/0.425 | Improved |
| add WTDen on top of STSync (ablation) | STSync · OPV2V | AP@0.5: 0.818 -> 0.834 | Improved |
| add AdpSel on top of STSync+WTDen (full CATNet) (ablation) | STSync + WTDen · DAIR-V2X | AP@0.7: 0.549 -> 0.565 | Improved |
| add WTDen on top of STSync+AdpSel (full CATNet) (ablation) | STSync + AdpSel · DAIR-V2X | AP@0.7: 0.553 -> 0.565 | Improved |
| vary AdpSel retention ratio from 0.1 to 0.6 (ablation) | standalone AdpSel · OPV2V | AP@0.7: 0.674 (0.1) / 0.691 (0.3) / 0.663 (0.6) | Mixed |
| mask high-attention regions selected by AdpSel (ablation) | AdpSel · OPV2V | AP@0.5: 0.897 -> 0.364 | Hurt |
| mask low-attention regions (ablation) | AdpSel · OPV2V | AP@0.5: 0.897 -> 0.784 | Hurt |
| CATNet vs second-best method under latency and noise | CATNet · V2XSet | AP@0.5 / AP@0.7: 0.817 -> 0.858 / 0.624 -> 0.643 | Improved |
| increase max delay from 0ms to 500ms | CATNet · OPV2V | AP@0.7: 0.763 -> 0.624 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- ERMVP — ERMVP: Communication-Efficient and Collaboration-Robust Multi-Vehicle Perception in Challenging Environments (CVPR 2024) · better on 2 of 2 shared results
- How2comm — How2comm: Communication-Efficient and Collaboration-Pragmatic Multi-Agent Perception (NeurIPS 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Multi-agent Collaborative Perception via Motion-aware Robust Communication Network (CVPR 2024 · similarity 0.87)
- An Extensible Framework for Open Heterogeneous Collaborative Perception (ICLR 2024 · similarity 0.87)
- What Makes Good Collaborative Views? Contrastive Mutual Information Maximization for Multi-Agent Perception (AAAI 2024 · similarity 0.86)
- CoPEFT: Fast Adaptation Framework for Multi-Agent Collaborative Perception with Parameter-Efficient Fine-Tuning (AAAI 2025 · similarity 0.86)
- NegoCollab: A Common Representation Negotiation Approach for Heterogeneous Collaborative Perception (NeurIPS 2025 · similarity 0.85)
- RCDN: Towards Robust Camera-Insensitivity Collaborative Perception via Dynamic Feature-based 3D Neural Modeling (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
