AI Paper Atlas

Research map / Video / Video segmentation and tracking

Hierarchical Contrastive Inconsistency Learning for Deepfake Video Detection

Zhihao Gu, Taiping Yao, Yang Chen, Shouhong Ding, Lizhuang Ma

ECCV 2022 · Video › Video segmentation and tracking · approach: Deepfake video detection · 47 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 RAI module to ResNet-50 baseline (ablation)ResNet-50 · FF++ NTaccuracy: 87.50% -> 92.14%Improved
add RAI module to ResNet-50 baseline (ablation)ResNet-50 · FF++ F2Faccuracy: 88.93% -> 93.21%Improved
add ISF module to ResNet-50 baseline (ablation)ResNet-50 · FF++ NTaccuracy: 0.8750 -> 0.9178Improved
add RAI and ISF together (ablation)ResNet-50 · FF++ NTaccuracy: 0.9214 -> 0.9250Improved
add hierarchical contrast losses Ll and Lg to baseline features (ablation)ResNet-50 · FF++ F2Faccuracy: 88.93% -> 91.43%Improved
add hierarchical contrast on top of RAI and ISF (all components) (ablation)HCIL · FF++ NTaccuracy: 0.9250 -> 0.9464Improved
add snippet-level contrast (vanilla NCE) (ablation)HCIL · FF++ F2Faccuracy: 0.9464 -> 0.9536Improved
use weighted NCE (beta) instead of vanilla NCE (ablation)HCIL · FF++ F2Faccuracy: 0.9536 -> 0.9571Improved
use weighted NCE instead of vanilla NCE (ablation)HCIL · Celeb-DFaccuracy: 0.9981 -> 0.9981No clear effect
use weighted NCE instead of vanilla NCE (ablation)HCIL · DFDC (partial forgery)accuracy: 0.9395 -> 0.9511Improved
add snippet contrast Ll on RAI vs baseline without snippet contrast (ablation)HCIL · Cross-dataset DFDCAUC: 0.6323 -> 0.6898Improved
add all components (RAI, ISF, Ll, Lg) (ablation)HCIL · Cross-dataset Celeb-DFAUC: 0.6956 -> 0.7900Improved

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