Research map / Vision-language and multimodal / Cross-modal learning
UMCL: Unimodal-generated Multimodal Contrastive Learning for Cross-compression-rate Deepfake Detection
Ching-Yi Lai, Chih-Yu Jian, Pei-Cheng Chuang, Chia-Ming Lee, Chih–Chung Hsu, Hsu Chiou-Ting, Chia‐Wen Lin
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 |
|---|---|---|---|
| replace CPML attention fusion with ASA and add CQSL (UMCL vs CPML) | CPML · FF++ raw -> c23 | AUC (%): 91.27 -> 95.69 | Improved |
| UMCL vs CPML under training on c23, testing on raw | CPML · FF++ c23 -> raw | AUC (%): 90.13 -> 96.46 | Improved |
| UMCL vs CPML in-compression on c23 | CPML · FF++ c23 -> c23 | AUC (%): 97.42 -> 96.59 | Hurt |
| UMCL vs CPML in-compression on c40 | CPML · FF++ c40 -> c40 | AUC (%): 91.36 -> 90.70 | Hurt |
| UMCL vs CPML on DFDC to FF++ raw transfer | CPML · DFDC -> FF++ (raw) | AUC (%): 80.77 -> 89.11 | Improved |
| UMCL vs CPML trained on F2F, cross-manipulation average | CPML · FF++ c23 trained on F2F | AUC (%): 94.16 -> 96.79 | Improved |
| UMCL vs CPML trained on NT, cross-manipulation average | CPML · FF++ c23 trained on NT | AUC (%): 94.85 -> 94.53 | No clear effect |
| UMCL vs CPML with rPPG sampling ratio 30% | CPML · FF++ (raw) train and test | AUC (%): 54.22 -> 86.49 | Improved |
| UMCL vs CPML with landmark perturbation sigma=0.05 | CPML · FF++ (raw) train and test | AUC (%): 34.40 -> 76.23 | Improved |
| UMCL vs CPML with no degradation on c40 test | CPML · FF++ (raw) -> c40 | AUC (%): 63.03 -> 79.72 | Improved |
| UMCL vs CPML in-domain DFDC | CPML · DFDC | AUC (%): 88.76 -> 92.17 | Improved |
Closest papers by meaning
- Standing on the Shoulders of Giants: Reprogramming Visual-Language Model for General Deepfake Detection (AAAI 2025 · similarity 0.87)
- Multi-View Collaborative Learning Network for Speech Deepfake Detection (AAAI 2025 · similarity 0.86)
- Multi-modal Deepfake Detection via Multi-task Audio-Visual Prompt Learning (AAAI 2025 · similarity 0.86)
- AVFF: Audio-Visual Feature Fusion for Video Deepfake Detection (CVPR 2024 · similarity 0.86)
- Generalizing Deepfake Video Detection with Plug-and-Play: Video-Level Blending and Spatiotemporal Adapter Tuning (CVPR 2025 · similarity 0.86)
- SIDA: Social Media Image Deepfake Detection, Localization and Explanation with Large Multimodal Model (CVPR 2025 · similarity 0.86)
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.
