AI Paper Atlas

Research map / Video / Video segmentation and tracking

Exploring Temporal Coherence for More General Video Face Forgery Detection

Yinglin Zheng, Jianmin Bao, Dong Chen, Ming Zeng, Fang Wen

ICCV 2021 · Video › Video segmentation and tracking · approach: Deepfake video detection · 352 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
set all spatial convolution kernels to 1 (FTCN) (ablation)3D R50 · FF++ HQ train on F2F, test avgvideo-level AUC (%): 90.3 -> 97.5Improved
use spatial-only convolutions instead of spatio-temporal (ablation)3D R50 · FF++ HQ train on F2F, test avgvideo-level AUC (%): 90.3 -> 77.7Hurt
add 3x3 spatial kernel in first FTCN conv layer (ablation)3D R50-FTCN · FF++ HQ train on F2F, test avgvideo-level AUC (%): 97.5 -> 96.8Hurt
add 5x5 spatial kernel in first FTCN conv layer (ablation)3D R50-FTCN · FF++ HQ train on F2F, test avgvideo-level AUC (%): 97.5 -> 95.1Hurt
spatially shuffle the input pixels of test/training clips (ablation)3D R50-FTCN · FF++ HQ train on F2F, test avgvideo-level AUC (%): 97.5 -> 95.8Hurt
use FTCN design instead of parameter-matched channel-reduced 3D R50 (ablation)3D R50-SP · FF++ HQ train on F2F, test avgvideo-level AUC (%): 89.6 -> 97.5Improved
use height-only (FHCN) convolutions instead of temporal-only (ablation)3D R50-FTCN · FF++ HQ train on F2F, test avgvideo-level AUC (%): 97.5 -> 77.1Hurt
use width-only (FWCN) convolutions instead of temporal-only (ablation)3D R50-FTCN · FF++ HQ train on F2F, test avgvideo-level AUC (%): 97.5 -> 83.4Hurt
increase clip size from 8 to 32 frames (ablation)3D R50-FTCN · FF++ HQ train on F2F, test avgvideo-level AUC (%): 87.7 -> 97.5Improved
increase clip size from 32 to 64 frames (ablation)3D R50-FTCN · FF++ HQ train on F2F, test avgvideo-level AUC (%): 97.5 -> 97.9No clear effect
add one-layer Temporal Transformer after FTCN (ablation)3D R50-FTCN · FF++ HQ train on F2F, test avgvideo-level AUC (%): 97.5 -> 98.9Improved
increase Temporal Transformer layers from 1 to 2 (ablation)FTCN+TT · FF++ HQ train on F2F, test avgvideo-level AUC (%): 98.9 -> 98.5Hurt

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