AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Self-Supervised Visual Acoustic Matching

Arjun Somayazulu, Changan Chen, Kristen Grauman

NeurIPS 2023 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Pretext-task visual pre-training · 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
add visual input (vs non-visual LeMARA) (ablation)LeMARA · SoundSpaces-Speech unseenRTE (s): 0.152 -> 0.079Improved
add visual input (vs non-visual LeMARA) (ablation)LeMARA · AVSpeech-Rooms unseenRTE (s): 0.137 -> 0.071Improved
add visual input (vs non-visual LeMARA) (ablation)LeMARA · LibriSpeechRTE (s): 0.223 -> 0.210Improved
use acoustic residue metric instead of SRMR alone as training objective (ablation)LeMARA · LibriSpeechRTE: 0.2308 -> 0.2156Improved
combine acoustic residue with SRMR (ablation)LeMARA · LibriSpeechRTE: 0.2156 -> 0.2123Improved
add shortcut training with SoundSpaces de-biaser (ablation)LeMARA · LibriSpeechRTE: 0.2123 -> 0.2100Improved
LeMARA training objective vs ViGAS with same reverberator architectureLeMARA · SoundSpaces-Speech unseenRTE (s): 0.108 -> 0.079Improved
LeMARA vs AViTAR on STFT errorLeMARA · AVSpeech-Rooms unseenSTFT: 2.894 -> 6.298Hurt
LeMARA vs AViTAR on logSTFTLeMARA · AVSpeech-Rooms unseenlogSTFT: 1.290 -> 0.571Improved
LeMARA vs AViTAR in human room identificationLeMARA · Human studyaccuracy (%): 34.7 -> 46.1Improved
de-biasing vs dereverberation of audiode-biaser · AVSpeech-RoomsSRMR: 8.99 -> 13.14Improved
de-biasing vs dereverberation of audiode-biaser · SoundSpaces-SpeechRT60: 0.06 -> 0.04Improved

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