AI Paper Atlas

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

Objects that Sound

Relja Arandjelovic, Andrew Zisserman

ECCV 2018 · Vision-language and multimodal › Cross-modal learning · approach: Audio-visual source localization · 452 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
compute correspondence from Euclidean distance between aligned embeddings (AVE-Net) instead of concatenation and FC layers (L3-Net)L3-Net · AudioSet-Instruments cross-modal retrieval (image-to-audio)nDCG@30: .418 -> .561Improved
same change as aboveL3-Net · AudioSet-Instruments cross-modal retrieval (audio-to-image)nDCG@30: .385 -> .587Improved
same change as aboveL3-Net · AudioSet-Instruments intra-modal retrieval (audio-to-audio)nDCG@30: .653 -> .665Improved
train AVE-Net from unlabelled video instead of using supervised ImageNet VGG16 featuresVGG16-ImageNet · AudioSet-Instruments image-to-imagenDCG@30: .600 -> .604Improved
align L3-Net features with CCAL3-Net · AudioSet-Instruments image-to-audionDCG@30: .418 -> .531Improved
align ImageNet VGG16 vision features with L3 audio features by CCA instead of learning embeddingsAVE-Net · AudioSet-Instruments image-to-audionDCG@30: .561 -> .458Hurt
use multi-frame (25 frames) visual input (ablation)AVE-Net · AudioSet-Instruments AVCAVC accuracy: 81.9% -> 84.7%Improved
use frame plus optical flow visual input (ablation)AVE-Net · AudioSet-Instruments AVCAVC accuracy: 81.9% -> 84.9%Improved
use frame plus optical flow visual input (ablation)AVE-Net · AudioSet-Instruments image-to-image retrievalnDCG@30: .604 -> .608No clear effect
use frame plus optical flow visual input (ablation)AVE-Net · AudioSet-Instruments image-to-audio retrievalnDCG@30: .561 -> .558No clear effect
add shortcut prevention by sampling negative audio only at multiples of 0.04s (ablation)AVE-Net · AudioSet-Instruments retrievalnDCG@30: 1-2% better retrievalImproved
add shortcut prevention by sampling negative audio only at multiples of 0.04s (ablation)AVE-Net · AudioSet-Instruments AVCAVC accuracy: 87.6% -> 81.9%Mixed

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