FaithfulFaces：面向文本生成视频的姿态保真人脸身份保持框架

View PDF HTML (experimental)

Abstract:Identity-preserving text-to-video generation (IPT2V) empowers users to produce diverse and imaginative videos with consistent human facial identity. Despite recent progress, existing methods often suffer from significant identity distortion under large facial pose variations or facial occlusions. In this paper, we propose FaithfulFaces, a pose-faithful facial identity preservation learning framework to improve IPT2V in complex dynamic scenes. The key of FaithfulFaces is a pose-shared identity aligner that refines and aligns facial poses across distinct views via a pose-shared dictionary and a pose variation-identity invariance constraint. By mapping single-view inputs into a global facial pose representation with explicit Euler angle embeddings, FaithfulFaces provides a pose-faithful facial prior that guides generative foundations toward robust identity-preserving generation. In particular, we develop a specialized pipeline to curate a high-quality video dataset featuring substantial facial pose diversity. Extensive experiments demonstrate that FaithfulFaces achieves state-of-the-art performance, maintaining superior identity consistency and structural clarity even as pose changes and occlusions occur. Project Page: this https URL

Comments:	Accepted by NeurIPS 2026
Subjects:	Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Cite as:	arXiv:2605.04702 [cs.CV]
	(or arXiv:2605.04702v2 [cs.CV] for this version)
	https://doi.org/10.48550/arXiv.2605.04702 arXiv-issued DOI via DataCite

Submission history

From: Yuanzhi Wang [view email]
[v1] Wed, 6 May 2026 09:54:09 UTC (17,129 KB)
[v2] Fri, 16 Oct 2026 17:56:38 UTC (17,139 KB)