Composing Driving Worlds through Disentangled Control for Adversarial Scenario Generation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914390963388416 |
|---|---|
| author | Zhan, Yifan Chen, Zhengqing Wang, Qingjie He, Zhuo Niu, Muyao Guo, Xiaoyang Yin, Wei Ren, Weiqiang Zhang, Qian Zheng, Yinqiang |
| author_facet | Zhan, Yifan Chen, Zhengqing Wang, Qingjie He, Zhuo Niu, Muyao Guo, Xiaoyang Yin, Wei Ren, Weiqiang Zhang, Qian Zheng, Yinqiang |
| contents | A major challenge in autonomous driving is the "long tail" of safety-critical edge cases, which often emerge from unusual combinations of common traffic elements. Synthesizing these scenarios is crucial, yet current controllable generative models provide incomplete or entangled guidance, preventing the independent manipulation of scene structure, object identity, and ego actions. We introduce CompoSIA, a compositional driving video simulator that disentangles these traffic factors, enabling fine-grained control over diverse adversarial driving scenarios. To support controllable identity replacement of scene elements, we propose a noise-level identity injection, allowing pose-agnostic identity generation across diverse element poses, all from a single reference image. Furthermore, a hierarchical dual-branch action control mechanism is introduced to improve action controllability. Such disentangled control enables adversarial scenario synthesis-systematically combining safe elements into dangerous configurations that entangled generators cannot produce. Extensive comparisons demonstrate superior controllable generation quality over state-of-the-art baselines, with a 17% improvement in FVD for identity editing and reductions of 30% and 47% in rotation and translation errors for action control. Furthermore, downstream stress-testing reveals substantial planner failures: across editing modalities, the average collision rate of 3s increases by 173%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_12864 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Composing Driving Worlds through Disentangled Control for Adversarial Scenario Generation Zhan, Yifan Chen, Zhengqing Wang, Qingjie He, Zhuo Niu, Muyao Guo, Xiaoyang Yin, Wei Ren, Weiqiang Zhang, Qian Zheng, Yinqiang Computer Vision and Pattern Recognition A major challenge in autonomous driving is the "long tail" of safety-critical edge cases, which often emerge from unusual combinations of common traffic elements. Synthesizing these scenarios is crucial, yet current controllable generative models provide incomplete or entangled guidance, preventing the independent manipulation of scene structure, object identity, and ego actions. We introduce CompoSIA, a compositional driving video simulator that disentangles these traffic factors, enabling fine-grained control over diverse adversarial driving scenarios. To support controllable identity replacement of scene elements, we propose a noise-level identity injection, allowing pose-agnostic identity generation across diverse element poses, all from a single reference image. Furthermore, a hierarchical dual-branch action control mechanism is introduced to improve action controllability. Such disentangled control enables adversarial scenario synthesis-systematically combining safe elements into dangerous configurations that entangled generators cannot produce. Extensive comparisons demonstrate superior controllable generation quality over state-of-the-art baselines, with a 17% improvement in FVD for identity editing and reductions of 30% and 47% in rotation and translation errors for action control. Furthermore, downstream stress-testing reveals substantial planner failures: across editing modalities, the average collision rate of 3s increases by 173%. |
| title | Composing Driving Worlds through Disentangled Control for Adversarial Scenario Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.12864 |