AI Paper Atlas

Research map / Video / Action recognition

Social Scene Understanding: End-To-End Multi-Person Action Localization and Collective Activity Recognition

Timur Bagautdinov, Alexandre Alahi, Francois Fleuret, Pascal Fua, Silvio Savarese

CVPR 2017 · Video › Action recognition · approach: Temporal action segmentation · 267 citations (OpenAlex)

Check 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 temporal matching RNN to the single-frame model (ablation)OURS-single on Volleyball · Volleyballcollective activity accuracy with MRF detections: 83.3 -> 87.1Improved
Add temporal matching RNN to the single-frame model (ablation)OURS-single on Volleyball · Volleyballindividual action accuracy with MRF detections: 77.8 -> 77.9No clear effect
Use ground-truth boxes rather than MRF detections (ablation)OURS-temporal on Volleyball · Volleyballindividual action accuracy: 77.9 -> 82.4Improved
Use embedding-space nearest-neighbor matching instead of box-coordinate matching (ablation)OURS-temporal with MRF detections · Volleyballcollective activity accuracy: 82.0 -> 87.1Improved
Use embedding-space nearest-neighbor matching instead of box-coordinate matching (ablation)OURS-temporal with MRF detections · Volleyballindividual action accuracy: 68.6 -> 77.9Improved
Use soft embedding matching instead of embedding nearest-neighbor matching (ablation)OURS-temporal with MRF detections · Volleyballcollective activity accuracy: 87.1 -> 86.2Hurt
Use soft embedding matching instead of embedding nearest-neighbor matching (ablation)OURS-temporal with ground-truth detections · Volleyballcollective activity accuracy: 90.0 -> 90.6Improved
Replace NMS with MRF detection refinement (ablation)OURS-temporal with boxes matching · Volleyballcollective activity accuracy: 77.0 -> 82.0Improved
Replace NMS with MRF detection refinement (ablation)OURS-temporal with embedding matching · Volleyballcollective activity accuracy: 85.2 -> 87.1Improved
Replace NMS with MRF detection refinement (ablation)OURS-temporal with soft embedding matching · Volleyballindividual action accuracy: 75.7 -> 77.4Improved
Use a shared multi-task representation rather than a person-only classifierOURS-single with ground-truth boxes versus Inception-person · Volleyballindividual action accuracy: 78.1 -> 81.1Improved

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