AI Paper Atlas

Research map / Continual learning / Online and replay-based

Continual Face Forgery Detection via Historical Distribution Preserving

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

IJCV 2024 · Continual learning › Online and replay-based · approach: Dynamic expansion and isolation · 18 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 pseudo-forged entropy training with UAP samples to the baseline (ablation)SFT · CFFD Protocol 1PRE ACC: 67.29 -> 74.33Improved
add pseudo-forged feature distillation on top of pseudo-forged training (ablation)HDP · CFFD Protocol 1PRE ACC: 74.33 -> 77.39Improved
add real feature distillation alone to the baseline (ablation)SFT · CFFD Protocol 1PRE ACC: 67.29 -> 81.19Improved
add real feature distillation to pseudo-forged training (ablation)HDP · CFFD Protocol 1PRE ACC: 74.33 -> 90.71Improved
add pseudo-forged feature distillation on top of the other two terms (ablation)HDP · CFFD Protocol 1PRE ACC: 90.71 -> 94.94Improved
lower UAP attack rate from 0.8 to 0.6 (ablation)HDP · CFFD Protocol 1PRE ACC: 94.94 -> 90.33Hurt
raise UAP attack rate from 0.8 to 1.0 (ablation)HDP · CFFD Protocol 1PRE ACC: 94.94 -> 93.67Hurt
add HDP to sequence fine-tuningSFT · CFFD Protocol 1AVG ACC: 75.07 -> 93.32Improved
add a 50-image per-stage buffer to HDPHDP · CFFD Protocol 2PRE ACC: 86.99 -> 88.57Improved
add a 50-image per-stage buffer to HDPHDP · CFFD Protocol 3AVG ACC: 77.17 -> 80.15Improved
add HDP to MAT detectorMAT · CFFD Protocol 1AVG ACC: 76.81 -> 93.13Improved
add HDP to F3-Net detectorF3-Net · CFFD Protocol 1AVG ACC: 77.18 -> 91.49Improved

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