FreeAction: Training-Free Techniques for Enhanced Fidelity of Trajectory-to-Video Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909814449242112 |
|---|---|
| author | Kim, Seungwook Lee, Seunghyeon Cho, Minsu |
| author_facet | Kim, Seungwook Lee, Seunghyeon Cho, Minsu |
| contents | Generating realistic robot videos from explicit action trajectories is a critical step toward building effective world models and robotics foundation models. We introduce two training-free, inference-time techniques that fully exploit explicit action parameters in diffusion-based robot video generation. Instead of treating action vectors as passive conditioning signals, our methods actively incorporate them to guide both the classifier-free guidance process and the initialization of Gaussian latents. First, action-scaled classifier-free guidance dynamically modulates guidance strength in proportion to action magnitude, enhancing controllability over motion intensity. Second, action-scaled noise truncation adjusts the distribution of initially sampled noise to better align with the desired motion dynamics. Experiments on real robot manipulation datasets demonstrate that these techniques significantly improve action coherence and visual quality across diverse robot environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_24241 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FreeAction: Training-Free Techniques for Enhanced Fidelity of Trajectory-to-Video Generation Kim, Seungwook Lee, Seunghyeon Cho, Minsu Computer Vision and Pattern Recognition Robotics Generating realistic robot videos from explicit action trajectories is a critical step toward building effective world models and robotics foundation models. We introduce two training-free, inference-time techniques that fully exploit explicit action parameters in diffusion-based robot video generation. Instead of treating action vectors as passive conditioning signals, our methods actively incorporate them to guide both the classifier-free guidance process and the initialization of Gaussian latents. First, action-scaled classifier-free guidance dynamically modulates guidance strength in proportion to action magnitude, enhancing controllability over motion intensity. Second, action-scaled noise truncation adjusts the distribution of initially sampled noise to better align with the desired motion dynamics. Experiments on real robot manipulation datasets demonstrate that these techniques significantly improve action coherence and visual quality across diverse robot environments. |
| title | FreeAction: Training-Free Techniques for Enhanced Fidelity of Trajectory-to-Video Generation |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2509.24241 |