AI Paper Atlas

Research map / Video / Video segmentation and tracking

Spherical World-Locking for Audio-Visual Localization in Egocentric Videos

Heeseung Yun, Ruohan Gao, Ishwarya Ananthabhotla, Anurag Kumar, Jacob Donley, Chao Li, Gunhee Kim, Vamsi Krishna Ithapu, Calvin Murdock

ECCV 2024 · Video › Video segmentation and tracking · approach: Layered video decomposition and dynamics · 6 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 MuST with spherical world-locking instead of AVLN/MAVASLegocentric active speaker localization · EasyCommAP: 86.32 -> 89.88Improved
add head pose information to MuST (ablation)MuST · EasyCommAP: 87.76 -> 89.88Improved
add rotation quaternion spatial similarity (ablation)MuST · EasyCommAP: 88.83 -> 89.88Improved
add pose information to audio-only spherical localization (ablation)MuST · RLR-CHATMAE g->p (degrees): 29.33 -> 15.23Improved
use multichannel instead of mono audio (ablation)MuST with pose and visual · EasyCommAP: 73.47 -> 89.88Improved
add visual input on top of multichannel audio and pose (ablation)MuST · EasyCommAP: 89.50 -> 89.88Improved
add modality-wise layer norm and attention projections (ablation)MuST · EasyCommAP: 88.53 -> 89.88Improved
add modality-wise MLP on top of M-LN and M-Attn variants (ablation)MuST · EasyCommAP: 89.67 -> 89.16Hurt
lengthen temporal window from 100ms to 300ms (ablation)MuST · EasyCommAP: 82.96 -> 89.88Improved
use sparse grid decoding instead of sparse point decoding (ablation)MuST · EasyCommAP: 88.95 -> 89.88Improved
jointly train all behavior anticipation tasks versus single-task (ablation)MuST AVB · AEA gaze 700msMAE: 15.02 -> 14.75Improved
use audio-visual input without previous behavior context for trajectory (ablation)MuST AV · AEA trajectory 300msMAE: 9.77 -> 92.38Hurt

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