AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

Intra-modal and Cross-modal Synchronization for Audio-visual Deepfake Detection and Temporal Localization

Ashutosh Anshul, Shreyas Gopal, Deepu Rajan, Eng Siong Chng

ICCV 2025 · Vision-language and multimodal › Cross-modal learning · approach: Self-supervised audio-visual representation learning · 2 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
use Gaussian-target MSE loss instead of InfoNCE (ablation)Decoder, VoxCeleb2 pretraining · FakeAVCelebAUC: 80.1 -> 88.0Improved
use Gaussian-target MSE loss instead of InfoNCE (ablation)Decoder, VoxCeleb2 pretraining · FakeAVCelebAP: 99.4 -> 99.7Improved
add intra-modal synchronization (AA, VV) pretraining (ablation)Decoder, VoxCeleb2 pretraining · FakeAVCelebAUC: 79.7 -> 88.0Improved
add intra-modal synchronization (AA, VV) pretraining (ablation)Decoder, VoxCeleb2 pretraining · FakeAVCelebAP: 99.4 -> 99.7Improved
use unimodal embeddings instead of intra-modal sync features in stage two (ablation)classification stage · FakeAVCelebAP/AUC: +0.1% AP, +3.8% AUCImproved
use sparse encoder instead of decoder attention variant (ablation)synchronization model · FakeAVCeleb and KoDFAP/AUC: 8 hrs -> 2 hrs per epochNo clear effect
use sparse encoder instead of decoder attention variant (ablation)Ours (VoxCeleb2) · FakeAVCeleb RVFAAP: 98.6 -> 96.5Hurt
pretrain on VoxCeleb2 instead of LRS2 (decoder) (ablation)Decoder variant · KoDFAP: 56.7 -> 98.9Improved
pretrain on VoxCeleb2 instead of LRS2 (sparse encoder) (ablation)Sparse Encoder variant · KoDFAP: 98.6 -> 94.4Hurt
use Decoder VoxCeleb2 model instead of AVFFdeepfake classification · FakeAVCeleb RVFAAP: 93.3 -> 98.6Improved
use Decoder VoxCeleb2 model instead of UMMAFormertemporal localization · LAV-DFAP@0.95: 37.61 -> 5.72Hurt

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