AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Information Bottleneck Disentanglement for Identity Swapping

Gege Gao, Huaibo Huang, Chaoyou Fu, Zhaoyang Li, Ran He

CVPR 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Fair and quality-aware face recognition · 124 citations (OpenAlex)

Check 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 the Informative Identity Bottleneck (IIB) (ablation)InfoSwap · FaceForensics++cosine similarity with source: 0.529 ± 0.118 -> 0.633 ± 0.098Improved
Add the Informative Identity Bottleneck (IIB) (ablation)InfoSwap · FaceForensics++ID retrieval (%): 96.3 -> 99.7Improved
Add Identity Contrastive Loss (ICL) rather than conventional identity loss (ablation)InfoSwap · FaceForensics++cosine similarity with target: 0.244 ± 0.118 -> 0.077 ± 0.104Improved
Add Identity Contrastive Loss (ICL) rather than conventional identity loss (ablation)InfoSwap · FaceForensics++cosine similarity with source: 0.544 ± 0.111 -> 0.633 ± 0.098Improved
Use the target information controller during integration (ablation)InfoSwap · FaceForensics++cosine similarity with source: 0.550 ± 0.110 -> 0.633 ± 0.098Improved
Use the target information controller during integration (ablation)InfoSwap · FaceForensics++ID retrieval (%): 98.5 -> 99.7Improved

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