AI Paper Atlas

Research map / Video / Action recognition

Learning Representations from Audio-Visual Spatial Alignment

Pedro Morgado, Yi Li, Nuno Nvasconcelos

NeurIPS 2020 · Video › Action recognition · approach: Contrastive video representation learning · 22 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
Pretrain with audio-visual spatial alignment instead of AVCR2+1D-18 audio-visual encoder · YT-360 AVSA-Binaccuracy: 61.20 -> 69.97Improved
Pretrain with AVSA instead of AVCR2+1D-18 · UCF101Video@1 accuracy: 69.68 -> 73.80Improved
Pretrain with AVSA instead of AVTSR2+1D-18 · HMDBVideo@1 accuracy: 35.89 -> 37.66Improved
Add curriculum learning (AVC phase first) (ablation)AVSA · YT-360 segmentation (context)mIoU: 34.30 -> 35.83Improved
Use transformer attention instead of MLP in translation networks (ablation)AVSA · YT-360 segmentation (context)mIoU: 34.35 -> 35.83Improved
Use full ambisonics instead of mono audio (ablation)AVSA · UCF101Clip@1 accuracy: 64.90 -> 68.52Improved
Use ambisonics instead of stereo audio (ablation)AVSA · YT-360 AVSA-Binaccuracy: 71.11 -> 69.97Hurt
Increase number of viewpoints from 4 to 8 (ablation)AVSA · Semantic segmentation and UCFmIoU / accuracy: 35.83 -> 34.84 mIoU; 68.52 -> 66.44 UCFHurt
Add spatial negatives to instance negatives (ablation)AVSA · UCF101Clip@1 accuracy: 64.09 -> 68.52Improved
Add temporal negatives on top of spatial (ablation)AVSA · Segmentation vs UCFmIoU / accuracy: 35.83 -> 36.11 mIoU; 68.52 -> 65.77 UCFMixed
Train on hard negatives only (no easy-negative curriculum) (ablation)AVSA · YT-360 segmentationmIoU: 35.83 -> 20.97Hurt
Use transformer depth 2 instead of direct prediction (ablation)AVSA · UCF101Clip@1 accuracy: 65.59 -> 68.52Improved

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