A two-stage privileged context distillation framework, which transfers full context behavior from a dense teacher through consistency distillation and then refines deployment rollouts with distribution matching, is proposed, which transfers full context behavior from a dense teacher through consistency distillation.
Xu Guo, Zhengxuan Wei, Xinghui Li et al.· 0 citations
This work proposes TARS, a 3D-free video re-shooting paradigm that provides more accurate and temporally consistent camera control than prior methods, and introduces self-supervised training to learn camera dynamics and fundamental visual representations without paired re-shooting data or 3D reconstruction.
Jiwen Liu, Shujuan Li, Xiaohan Li et al.· arXiv.org· 0 citations
Vera is proposed, a unified human-centric S2V framework for single- and multi- person generation that improves human identity consistency, multi-person subject binding, and motion naturalness, while reducing identity confusion and excessive reference-image copying.
Yulong Xu, Xinyue Liu, Shujuan Li et al.· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
