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
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add SIA module to EfficientNet-B4 | EfficientNet-B4 · FF++ LQ | ACC: 86.95 -> 90.23 | Improved |
| add SIA module to EfficientNet-B0 | EfficientNet-B0 · FF++ LQ | ACC: 85.66 -> 87.03 | Improved |
| add SIA module to XceptionNet | XceptionNet · FF++ LQ | ACC: 86.86 -> 87.30 | Improved |
| add SIA module to MobileNet-v2 | MobileNet-v2 · FF++ LQ | ACC: 85.19 -> 86.78 | Improved |
| add SIA module to MobileNet-v2 | MobileNet-v2 · FF++ HQ | ACC: 95.58 -> 97.05 | Improved |
| replace SE-layer attention with SIA | EfficientNet-B4 · FF++ LQ | ACC: 89.05 -> 90.23 | Improved |
| replace non-local attention with SIA | EfficientNet-B4 · FF++ LQ | ACC: 89.55 -> 90.23 | Improved |
| replace GSA attention with SIA | EfficientNet-B4 · FF++ HQ | ACC: 94.68 -> 97.64 | Improved |
| insert SIA at layers L1,L2,L4 instead of L3,L4,L5 (ablation) | SIA on EfficientNet-B4 · FF++ LQ | ACC: 89.14 -> 90.23 | Improved |
| insert SIA at layers L1,L2,L4 instead of L1,L4,L6 (ablation) | SIA on EfficientNet-B4 · FF++ LQ | AUC: 90.21 -> 93.75 | Improved |
| insert SIA at L1,L2,L4 instead of L1,L2,L3 (ablation) | SIA on EfficientNet-B4 · FF++ LQ | ACC: 89.56 -> 90.23 | Improved |
| use SIA instead of SPSL for cross-dataset generalization | EfficientNet-B4 · Celeb-DF (train FF++ LQ) | AUC: 76.88 -> 77.35 | Improved |
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.
- MobileNet-v2 — MobileNetV2: Inverted Residuals and Linear Bottlenecks (CVPR 2018) · better on 1 of 1 shared result
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
- Rethinking Vision-Language Model in Face Forensics: Multi-Modal Interpretable Forged Face Detector (CVPR 2025 · similarity 0.91)
- DF40: Toward Next-Generation Deepfake Detection (NeurIPS 2024 · similarity 0.89)
- Identity-Aware Vision-Language Model for Explainable Face Forgery Detection (AAAI 2026 · similarity 0.89)
- Critical Forgetting-Based Multi-Scale Disentanglement for Deepfake Detection (AAAI 2025 · similarity 0.89)
- Exposing the Deception: Uncovering More Forgery Clues for Deepfake Detection (AAAI 2024 · similarity 0.88)
- A Hitchhiker's Guide to Fine-Grained Face Forgery Detection Using Common Sense Reasoning (NeurIPS 2024 · similarity 0.88)
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.
