AI Paper Atlas

Research map / Video / Video generation

Towards Scene-Aware Video-to-Spatial Audio Generation

Jaeyeon Kim, Heeseung Yun, Gunhee Kim

IJCV 2026 · Video › Video generation · approach: Video-to-audio synthesis · 5 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
use end-to-end ViSAGe (directional) instead of Diff-Foley + Audio Spatial. two-stageViSAGe · YT-Ambigen testCC (all): 0.577 -> 0.721Improved
add visual guidance to directional guidance (ablation)ViSAGe · YT-Ambigen testFAD_dec: 5.56 -> 3.86Improved
add visual guidance to directional guidance (ablation)ViSAGe · YT-Ambigen testCC (all): 0.721 -> 0.635Hurt
add direction embedding (ablation)ViSAGe (PT) · YT-Ambigen testCC (all): 0.430 -> 0.524Improved
add rotation augmentation (ablation)ViSAGe · YT-Ambigen testCC (all): 0.531 -> 0.635Improved
add patchwise energy map (ablation)ViSAGe (PT+DIR) · YT-Ambigen testCC (all): 0.524 vs 0.531Mixed
add VGGSound pretraining (ablation)ViSAGe (DIR+PE) · YT-Ambigen testFAD_dec: 4.44 -> 4.01Improved
use proposed code pattern instead of sequential delay (ablation)ViSAGe · YT-Ambigen testFAD_dec: 10.75 -> 4.44Improved
use proposed pattern instead of residual only (ablation)ViSAGe · YT-Ambigen testCC (all): 0.480 -> 0.544Improved
use proposed pattern instead of spatial only (ablation)ViSAGe · YT-Ambigen testFAD_dec: 5.89 -> 4.44Improved
joint directional+visual guidance instead of dual 7.5 & 2.5 (ablation)ViSAGe · YT-Ambigen testCC (all): 0.596 -> 0.635Improved
train on YT-Ambigen instead of YT360video-to-audio model · dataset comparisonFAD: 15.91 -> 3.95Improved

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