AI Paper Atlas

Research map / Detection and segmentation / CNN features

AOT: Appearance Optimal Transport Based Identity Swapping for Forgery Detection

Hao Zhu, Chaoyou Fu, Qianyi Wu, Wayne Wu, Chen Qian, Ran He

NeurIPS 2020 · Detection and segmentation › CNN features · approach: Forgery detection · 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 3D features (PNCC and normals) to the transport (ablation)AOT · DPF-1.0SSIM-whole: 0.7183 -> 0.7810Improved
use NOTPE feature transfer instead of pixel-wise Add (ablation)AOT · DPF-1.0Gram loss: 0.036263 -> 0.003578Improved
use NOTPE instead of AdaIN (ablation)AOT · DPF-1.0SSIM-whole: 0.7522 -> 0.7810Improved
add the mix-and-segment discriminator (ablation)AOT · DPF-1.0SSIM-whole: 0.7255 -> 0.7810Improved
use MSD instead of the original discriminator (ablation)AOT · DPF-1.0Gram loss: 0.019354 -> 0.003578Improved
apply AOT appearance transfer on top of DFLDFL · FF++SSIM-whole: 0.7331 -> 0.8022Improved
apply AOT appearance transfer on top of FSGANFSGAN · DPF-1.0SSIM-edge: 0.7487 -> 0.7446Mixed
apply AOT on top of DFL for user judgementDFL · DPF-1.0 user studyRealism preference rate: 0.149041 -> 0.548493Improved
add 100 AOT-generated manipulated videos to forgery detector trainingI3D and TSN detectors · DPF-1.0 hidden setdetection accuracyImproved

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