AI Paper Atlas

Research map / Video / Video generation

TIGER: Taming Identity, Geometry, and Generative Priors for High-Quality Face Video Restoration

Yang Zhou, Wenxue Li, Peng Zhang, Yifei Chen, Fei Wang, Daiguo Zhou

ECCV 2026 · Video › Video generation · approach: Diffusion-based video editing · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use TIGER instead of DOVE (strongest baseline)TIGER · VFHQ-TestFVD: 103.892 -> 40.4222Improved
use TIGER instead of DOVETIGER · VFHQ-TestLPIPS: 0.1189 -> 0.0697Improved
use TIGER instead of DOVETIGER · VFHQ-TestIDS: 0.9283 -> 0.9481Improved
use TIGER instead of STARTIGER · VFHQ-TestCLIP-IQA: 0.5889 -> 0.4756Hurt
add identity prior (identity embedding input) (ablation)TIGER · VFHQ-TestVIRD, LPIPS, FVDImproved
add 3D geometry prior (normal maps) (ablation)TIGER · VFHQ-TestVIRD, LPIPS, FVDImproved
add Stage II decoder-coupled detail refinement (ablation)TIGER · VFHQ-Testperceptual qualityImproved
add Stage III adversarial distribution alignment (ablation)TIGER · VFHQ-Testidentity consistency, perceptual qualityImproved
use cross-source fusion (identity from reference, motion from degraded video) instead of single-source parameters (ablation)TIGER · VFHQ-TestFVR metricsImproved
add identity embeddings in cross-attention versus text-only (ablation)TIGER · VFHQ-Testoverall metricsImproved
increase latent loss weight lambda1 from 1.0 to 2.0 (ablation)TIGER · VoxCeleb2 (loss weight ablation)VIRD: 0.9280 -> 0.9315Hurt
reduce pixel loss weight lambda3 from 10.0 to 5.0 (ablation)TIGER · VoxCeleb2 (loss weight ablation)VIRD: 0.9280 -> 0.9356Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy