FaithfulFaces: Pose-Faithful Facial Identity Preservation for Text-to-Video Generation

TL;DR

FaithfulFaces introduces a pose-faithful learning framework that significantly enhances identity preservation in text-to-video generation, especially under pose variations and occlusions.

Contribution

The paper proposes a novel pose-shared identity aligner and a pose variation-identity invariance constraint for improved facial identity consistency in dynamic scenes.

Findings

Achieves state-of-the-art identity preservation under pose changes.

Maintains structural clarity despite facial occlusions.

Develops a high-quality dataset with diverse facial poses.

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 \textit{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.