AI Paper Atlas

Research map / Image generation / Editing and style transfer

FlowFace: Rectifying Identity Conditioning with Riemannian Geometry for Face Generation

Xinran Deng, Yiling Wu, Ye Tian, Libo Zhang

ECCV 2026 · Image generation › Editing and style transfer · approach: Diffusion face generation and 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
add dual-stream manifold encoder (ablation)FlowFace · FFHQ/CelebA-HQ identity-cluster benchmarkSimTarget: 48.3 -> 64.8Improved
add Lie-bracket (BCH) fusion (ablation)FlowFace · FFHQ/CelebA-HQ identity-cluster benchmarkSimTarget: 52.1 -> 64.8Improved
add geodesic regularization (ablation)FlowFace · FFHQ/CelebA-HQ identity-cluster benchmarkSimTarget: 55.7 -> 64.8Improved
add directional covariant regularization (ablation)FlowFace · FFHQ/CelebA-HQ identity-cluster benchmarkSimTarget: 58.4 -> 64.8Improved
add dual-stream manifold encoder (ablation)FlowFace · FFHQ/CelebA-HQ identity-cluster benchmarkFID: 95.4 -> 76.8Improved
add Lie-bracket fusion (ablation)FlowFace · FFHQ/CelebA-HQ identity-cluster benchmarkExpr.Acc: 42.6 -> 52.4Improved
add BCH interaction term to composition (ablation)FlowFace · Two-edit compositionOrderGap-ID: 0.146 -> 0.052Improved
add BCH interaction term (ablation)FlowFace · Three-edit compositionAvg. PermGap-ID: 0.182 -> 0.071Improved
BCH fusion instead of vector additionFlowFace · FFHQ/CelebA-HQ identity-cluster benchmarkSimTarget: 56.4 -> 64.8Improved
BCH fusion instead of attention weightingFlowFace · FFHQ/CelebA-HQ identity-cluster benchmarkSimTarget: 61.6 -> 64.8Improved
learned SPD metric training instead of Euclidean interpolationFlowFace · Trajectory stabilityCollapse Rate: 12.3% -> 1.2%Improved
learned SPD metric instead of hypersphere projectionFlowFace · Trajectory stabilitySmoothness: 0.89 -> 0.96Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 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