Search papers, labs, and topics across Lattice.
The paper introduces TIGER, a tri-prior fusion framework designed for high-quality face video restoration (FVR) that effectively addresses identity preservation, viewpoint entanglement, and perceptual realism. By incorporating subject-discriminative embeddings for identity anchoring and lifting 2D cues into a 3D parameter space for geometric guidance, TIGER ensures temporal consistency and structural fidelity across frames. Experimental results show that TIGER outperforms existing methods in both identity fidelity and temporal stability, marking a significant advancement in the field of FVR.
Achieving state-of-the-art identity fidelity and temporal stability, TIGER redefines the standards for high-quality face video restoration.
Face Video Restoration (FVR) aims to recover high-fidelity facial videos from degraded input while preserving identity and semantic consistency across frames. Existing methods often struggle to simultaneously address three key challenges: identity shift, viewpoint-entangled guidance, and perceptual realism. To tackle these issues, we propose TIGER, a structured tri-prior fusion framework that Tames Identity, Geometry, and gEnerative pRiors for high-quality FVR. Specifically, an Identity Prior is first established by injecting subject-discriminative embeddings into the latent space, effectively anchoring the subject's identity against severe degradations. Then, to provide temporally consistent structural guidance for dynamic videos, TIGER constructs a Geometry Prior by lifting 2D reference cues into a disentangled 3D parameter space, creating a geometric anchor through cross-source parameter fusion. Moreover, to achieve maximum efficiency without compromising realism, we harness the video generation model's Generative Prior through a one-step rectified flow. We further design a progressive three-stage training optimization strategy that refines structural fidelity, textural reconstruction, and distribution-level realism to ensure robust optimization. We also construct a large-scale FVR dataset to facilitate robust training and standardized evaluation. Extensive experiments demonstrate that TIGER achieves state-of-the-art performance in both identity fidelity and temporal stability, delivering a high-quality, efficient and identity-consistent FVR. Project page: https://yzhoulv.github.io/Tiger/.
