AI Paper Atlas

Research map / Detection and segmentation / CNN features

An Information Theoretic Approach for Attention-Driven Face Forgery Detection

Ke Sun, Hong Liu, Taiping Yao, Xiaoshuai Sun, Shen Chen, Shouhong Ding, Rongrong Ji

ECCV 2022 · Detection and segmentation › CNN features · approach: Forgery detection · 55 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 SIA module to EfficientNet-B4EfficientNet-B4 · FF++ LQACC: 86.95 -> 90.23Improved
add SIA module to EfficientNet-B0EfficientNet-B0 · FF++ LQACC: 85.66 -> 87.03Improved
add SIA module to XceptionNetXceptionNet · FF++ LQACC: 86.86 -> 87.30Improved
add SIA module to MobileNet-v2MobileNet-v2 · FF++ LQACC: 85.19 -> 86.78Improved
add SIA module to MobileNet-v2MobileNet-v2 · FF++ HQACC: 95.58 -> 97.05Improved
replace SE-layer attention with SIAEfficientNet-B4 · FF++ LQACC: 89.05 -> 90.23Improved
replace non-local attention with SIAEfficientNet-B4 · FF++ LQACC: 89.55 -> 90.23Improved
replace GSA attention with SIAEfficientNet-B4 · FF++ HQACC: 94.68 -> 97.64Improved
insert SIA at layers L1,L2,L4 instead of L3,L4,L5 (ablation)SIA on EfficientNet-B4 · FF++ LQACC: 89.14 -> 90.23Improved
insert SIA at layers L1,L2,L4 instead of L1,L4,L6 (ablation)SIA on EfficientNet-B4 · FF++ LQAUC: 90.21 -> 93.75Improved
insert SIA at L1,L2,L4 instead of L1,L2,L3 (ablation)SIA on EfficientNet-B4 · FF++ LQACC: 89.56 -> 90.23Improved
use SIA instead of SPSL for cross-dataset generalizationEfficientNet-B4 · Celeb-DF (train FF++ LQ)AUC: 76.88 -> 77.35Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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