Research map / 3D vision / Human pose estimation
Inferring Shared Attention in Social Scene Videos
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| add temporal optimization to Gaze+RP (ablation) | shared-attention detection model · VideoCoAtt test | Prediction Accuracy: 68.5 % -> 71.4 % | Helped |
| add temporal optimization to Gaze+RP (ablation) | shared-attention detection model · VideoCoAtt test | AUC: 0.747 -> 0.805 | Helped |
| add gaze estimation to the region-proposal model (ablation) | shared-attention detection model · VideoCoAtt test | Prediction Accuracy: 58.0 % -> 68.5 % | Helped |
| add region proposals to the gaze-only model (ablation) | shared-attention detection model · VideoCoAtt test | Prediction Accuracy: 64.0 % -> 68.5 % | Helped |
| add gaze estimation to the region-proposal model (ablation) | shared-attention detection model · VideoCoAtt test | AUC: 0.612 -> 0.747 | Helped |
| replace region proposals with saliency while retaining gaze and temporal optimization | shared-attention detection model · VideoCoAtt test | Prediction Accuracy: 71.4 % -> 66.2 % | Hurt |
| replace region proposals with saliency while retaining gaze and temporal optimization | shared-attention detection model · VideoCoAtt test | AUC: 0.805 -> 0.630 | Hurt |
| add raw image features and omit temporal optimization (ablation) | Gaze+RP model · VideoCoAtt test | Prediction Accuracy: 68.5 % -> 54.0 % | Hurt |
| use raw image input instead of explicit gaze and region proposal features (ablation) | shared-attention detection model · VideoCoAtt test | Prediction Accuracy: 71.4 % -> 52.3 % | Hurt |
Closest papers by meaning
- Attentional Push: A Deep Convolutional Network for Augmenting Image Salience With Shared Attention Modeling in Social Scenes (CVPR 2017 · similarity 0.90)
- Recurrent Mixture Density Network for Spatiotemporal Visual Attention (ICLR 2017 · similarity 0.88)
- Behavioral Geometric Supervision Aligns Video Foundation Models with Human Social Perception (NeurIPS 2026 · similarity 0.87)
- MTGS: A Novel Framework for Multi-Person Temporal Gaze Following and Social Gaze Prediction (NeurIPS 2024 · similarity 0.86)
- Multi-speaker Attention Alignment for Multimodal Social Interaction (CVPR 2026 · similarity 0.85)
- A Motion-aware Spatio-temporal Graph for Video Salient Object Ranking (NeurIPS 2024 · similarity 0.85)
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.