AI Paper Atlas

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

Audio-Visual Scene Analysis with Self-Supervised Multisensory Features

Andrew Owens, Alexei A. Efros

ECCV 2018 · Vision-language and multimodal › Cross-modal learning · approach: Self-supervised audio-visual representation learning · 722 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
add self-supervised audio-visual pretraining (ablation)Multisensory network · UCF-101 split 1 action recognitionaccuracy (%): 68.1% -> 82.1%Improved
add audio input to the pretrained network (ablation)Multisensory network · UCF-101 split 1 action recognitionaccuracy (%): 77.6% -> 82.1%Improved
use raw waveform instead of spectrogram audio (ablation)Multisensory network · UCF-101 split 1 action recognitionaccuracy (%): 81.1% -> 82.1%No clear effect
use misalignment pretext task instead of random audio-video pairing (ablation)Multisensory network · UCF-101 split 1 action recognitionaccuracy (%): 78.7% -> 82.1%Improved
use multisensory features instead of I3D Kinetics features (ablation)On/off separation model · VoxCeleb separation (all)On/off L1 error: 12.3 -> 11.4Improved
use multiple frames instead of a single frame (ablation)On/off separation model · VoxCeleb separation (all)On/off L1 error: 14.8 -> 11.4Improved
add the audio stream (early fusion) to the multisensory network (ablation)On/off separation model · VoxCeleb separation (all)On/off L1 error: 11.6 -> 11.4Improved
use stereo instead of mono audio (ablation)On/off separation model · VoxCeleb separation (all)On/off L1 error: 11.4 -> 11.4No clear effect
add self-supervised pretraining versus scratch (ablation)On/off separation model · VoxCeleb separation (all)On/off L1 error: 12.9 -> 11.4Improved
add permutation invariant (PIT) loss variantOn/off separation model · VoxCeleb separation (all)SDR: 7.0 -> 7.6Improved
use audio-visual model instead of audio-only u-net PITOn/off + PIT · VoxCeleb separation (all)SDR: 7.3 -> 7.6Improved
use ours instead of Hou et al. on short clipsOurs (on/off) · VoxCeleb short videos (200ms)On-SDR: 4.5 -> 7.6Improved

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