Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.16973 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915316774207488 |
|---|---|
| author | Jiang, Wentao Wang, Jingya Ji, Kaiyang Jia, Baoxiong Huang, Siyuan Shi, Ye |
| author_facet | Jiang, Wentao Wang, Jingya Ji, Kaiyang Jia, Baoxiong Huang, Siyuan Shi, Ye |
| contents | Human action-reaction synthesis, a fundamental challenge in modeling causal human interactions, plays a critical role in applications ranging from virtual reality to social robotics. While diffusion-based models have demonstrated promising performance, they exhibit two key limitations for interaction synthesis: reliance on complex noise-to-reaction generators with intricate conditional mechanisms, and frequent physical violations in generated motions. To address these issues, we propose Action-Reaction Flow Matching (ARFlow), a novel framework that establishes direct action-to-reaction mappings, eliminating the need for complex conditional mechanisms. Our approach introduces a physical guidance mechanism specifically designed for Flow Matching (FM) that effectively prevents body penetration artifacts during sampling. Moreover, we discover the bias of traditional flow matching sampling algorithm and employ a reprojection method to revise the sampling direction of FM. To further enhance the reaction diversity, we incorporate randomness into the sampling process. Extensive experiments on NTU120, Chi3D and InterHuman datasets demonstrate that ARFlow not only outperforms existing methods in terms of Fréchet Inception Distance and motion diversity but also significantly reduces body collisions, as measured by our new Intersection Volume and Intersection Frequency metrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_16973 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ARFlow: Human Action-Reaction Flow Matching with Physical Guidance Jiang, Wentao Wang, Jingya Ji, Kaiyang Jia, Baoxiong Huang, Siyuan Shi, Ye Computer Vision and Pattern Recognition Artificial Intelligence Human action-reaction synthesis, a fundamental challenge in modeling causal human interactions, plays a critical role in applications ranging from virtual reality to social robotics. While diffusion-based models have demonstrated promising performance, they exhibit two key limitations for interaction synthesis: reliance on complex noise-to-reaction generators with intricate conditional mechanisms, and frequent physical violations in generated motions. To address these issues, we propose Action-Reaction Flow Matching (ARFlow), a novel framework that establishes direct action-to-reaction mappings, eliminating the need for complex conditional mechanisms. Our approach introduces a physical guidance mechanism specifically designed for Flow Matching (FM) that effectively prevents body penetration artifacts during sampling. Moreover, we discover the bias of traditional flow matching sampling algorithm and employ a reprojection method to revise the sampling direction of FM. To further enhance the reaction diversity, we incorporate randomness into the sampling process. Extensive experiments on NTU120, Chi3D and InterHuman datasets demonstrate that ARFlow not only outperforms existing methods in terms of Fréchet Inception Distance and motion diversity but also significantly reduces body collisions, as measured by our new Intersection Volume and Intersection Frequency metrics. |
| title | ARFlow: Human Action-Reaction Flow Matching with Physical Guidance |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2503.16973 |