Reproduce “Chronos: Sequential Identity-Preserving Video Generation by Temporal Anchoring” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Mengqi Huang, Wenchuan Wang, Zhendong Mao, Yijing Tu, Zhike Liu, Luoni Wang, Ning Tong, Xiaoxiao Sheng and 3 more

Chronos generates videos from a person’s reference image and an action prompt while aiming to preserve identity and follow the requested motion. It refines prompts, creates identity-consistent visual frame anchors, and routes candidate videos using multiple evaluation criteria. On the sequential-action challenge, it reports a Total Score of 0.6102, 6.6% above Phantom-14B; this evidence is limited to the challenge setup.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.