AI Paper Atlas

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

Siamese Vision Transformers are Scalable Audio-visual Learners

Yan-Bo Lin, Gedas Bertasius

ECCV 2024 · Vision-language and multimodal › Cross-modal learning · approach: Self-supervised audio-visual representation learning · 9 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
share one ViT between audio and visual instead of separate encoders (ablation)AVSiam-Large · AudioSet-2MA+V mAP: 52.0 -> 52.1 with 640M -> 332M params, 20.6G -> 10.9G memoryNo clear effect
use multi-ratio masking instead of best fixed-ratio masking (ablation)AVSiam-Base · AudioSet-20KmAP: 40.8 -> 41.3 mAP; 362 -> 160 V100 hoursImproved
use multi-ratio masking instead of no-masking contrastive learning (ablation)AVSiam-Base · AudioSet-20KmAP: 40.4 -> 41.3 mAP; 510 -> 160 V100 hoursImproved
use higher fixed masking ratio (75%/75%) (ablation)AVSiam-Base · AudioSet-20KmAP: 40.8 -> 38.6 mAPHurt
use AST instead of ViT as shared encoder (ablation)AVSiam · AudioSet-20K / VGGSoundvisual-only and A+V: audio 36.5 -> 37.5; visual 23.7 -> 9.6; A+V 41.6 -> 38.1 (AS-20K)Mixed
add unsupervised audio-visual contrastive pretraining (ablation)AVSiam-Base · AudioSet-20K / VGGSoundmAP / accuracy: 39.9 -> 41.3 mAP; 61.2 -> 64.4 accImproved
add MAE reconstruction loss to contrastive pretraining (ablation)AVSiam-Base · AudioSet-20K / VGGSoundmAP / accuracy: 41.3 -> 41.6 mAP; 64.4 -> 64.9 acc; +17 V100 hoursImproved
scale pretraining data (add VGGSound and ACAV)AVSiam-Base · AS-20K / AS-2M / VGGSoundmAP / accuracy: +1.4, +1.3, +1.8Improved
scale backbone from Base to HugeAVSiam · AS-20K / AS-2M / VGGSoundmAP / accuracy: +3.4 mAP, +4.1 mAP, +3.1%Improved
use shared encoder instead of separate encoders on audio-only and visual-only inputsAVSiam vs CAV-MAE · AudioSet-2M visual-onlymAP: 11.1 -> 30.8Improved

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