Research map / Multi-agent systems and games / Human-AI and goal-conditioned
V2XPnP: Vehicle-to-Everything Spatio-Temporal Fusion for Multi-Agent Perception and Prediction
Zewei Zhou, Hao Xiang, Zhaoliang Zheng, Seth Z. Zhao, Mingyue Lei, Yun Zhang, Tianhui Cai, Xinyi Liu, Johnson Liu, Maheswari Bajji, Xin Xia, Zhiyu Huang and 2 more
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 |
|---|---|---|---|
| use one-step instead of multi-step communication | V2XPnP · V2XPnP-Seq-VC | EPA: 43.0 -> 48.2 | Improved |
| use one-step instead of multi-step communication | V2XPnP · V2XPnP-Seq-VC | AP@0.5: 68.2 -> 71.6 | Improved |
| add temporal fusion module (ablation) | V2XPnP · V2XPnP-Seq-VC | AP@0.5: 43.9 -> 57.2 | Improved |
| add multi-agent spatial fusion (ablation) | V2XPnP · V2XPnP-Seq-VC | EPA: 33.8 -> 44.4 | Improved |
| add map fusion (ablation) | V2XPnP · V2XPnP-Seq-VC | EPA: 44.4 -> 48.2 | Improved |
| add map fusion (ablation) | V2XPnP · V2XPnP-Seq-VC | FDE: 2.70 -> 2.36 | Improved |
| add map fusion (ablation) | V2XPnP · V2XPnP-Seq-VC | AP@0.5: 71.3 -> 71.6 | No clear effect |
| use V2XPnP temporal attention instead of FaF convolutions | intermediate fusion · V2XPnP-Seq-VC (temporal perception) | AP@0.5: 70.3 -> 74.0 | Improved |
| add temporal fusion (FaF) to no-fusion model | No Fusion · V2XPnP-Seq-VC | AP@0.5: 43.9 -> 53.4 | Improved |
| use intermediate V2XPnP instead of early fusion | fusion strategy · V2XPnP-Seq-VC | EPA: 36.7 -> 48.2 | Improved |
| use end-to-end FaF instead of decoupled no-fusion | No Fusion · V2XPnP-Seq-VC | EPA: 24.3 -> 31.6 | Improved |
| V2XPnP vs V2X-ViT in V2V mode | V2XPnP · V2XPnP-Seq-V2V | AP@0.5: 64.6 -> 70.5 | Improved |
Closest papers by meaning
- V2X-Real: a Largs-Scale Dataset for Vehicle-to-Everything Cooperative Perception (ECCV 2024 · similarity 0.91)
- End-to-End Autonomous Driving Through V2X Cooperation (AAAI 2025 · similarity 0.89)
- Privacy-Preserving V2X Collaborative Perception Integrating Unknown Collaborators (AAAI 2025 · similarity 0.89)
- V2X-Radar: A Multi-modal Dataset with 4D Radar for Cooperative Perception (NeurIPS 2025 · similarity 0.87)
- UrbanIng-V2X: A Large-Scale Multi-Vehicle, Multi-Infrastructure Dataset Across Multiple Intersections for Cooperative Perception (NeurIPS 2025 · similarity 0.87)
- One is Plenty: A Polymorphic Feature Interpreter for Immutable Heterogeneous Collaborative Perception (CVPR 2025 · similarity 0.87)
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.
