AI Paper Atlas

Research map / Video / Video generation

CanonSwap: High-Fidelity and Consistent Video Face Swapping via Canonical Space Modulation

Xiangyang Luo, Ye Zhu, Yunfei Liu, Lijian Lin, Cong Wan, Zijian Cai, Yu Li, Shao-Lun Huang

ICCV 2025 · Video › Video generation · approach: Diffusion-based human image animation · 16 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
add canonical-space warping before swapping (ablation)CanonSwap · VFS benchmarkPose error: 0.0227 -> 0.0159Improved
add canonical-space warping before swapping (ablation)CanonSwap · VFS benchmarkExpression error: 0.9512 -> 0.7592Improved
add the refinement module before warping back (ablation)CanonSwap · VFS benchmarkFVD: 481.77 -> 125.30Improved
add the refinement module before warping back (ablation)CanonSwap · VFS benchmarkID similarity: 0.4778 -> 0.5748Improved
add the soft spatial mask (Partial Identity Modulation) instead of global modulation (ablation)CanonSwap · VFS benchmarkFVD: 165.17 -> 125.30Improved
add the soft spatial mask (ablation)CanonSwap · VFS benchmarkID retrieval: 97.38 -> 99.78Improved
add canonical warping (ablation)CanonSwap · VFS benchmarkFVD: 131.57 -> 125.30Improved
use CanonSwap instead of strongest GAN baseline SimSwapVideo face swapping · VFS benchmarkLSE-C: 5.306 -> 6.053Improved
use CanonSwap instead of FSGAN on expressionVideo face swapping · FF++Expression error: 0.7184 -> 0.7328No clear effect
frame-by-frame canonical swap instead of video diffusionVideo face swapping · Inference speedFPS: 0.21 -> 14 (vs REFace)Improved

Later papers that compared with this method

1 later accepted paper reports CanonSwap as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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