Research map / Video / Action recognition
Spatio-Temporal Proximity-Aware Dual-Path Model for Panoramic Activity Recognition
Sumin Lee, Yooseung Wang, Sangmin Woo, Changick Kim
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add spatial (height and width) attention on top of temporal attention (ablation) | SPDP-Net individual self-attention · JRDB-PAR | Overall F1: 40.3 -> 46.5 | Improved |
| add any spatio-temporal attention versus none (ablation) | SPDP-Net · JRDB-PAR | Overall F1: 37.4 -> 46.5 | Improved |
| use K-means with predicted number of groups instead of spectral clustering (ablation) | SPDP-Net social group detection · JRDB-PAR | IoU@0.5: 49.1 -> 56.4 | Improved |
| add auxiliary loss L_aux on individual attention (ablation) | SPDP-Net · JRDB-PAR | Social group detection IoU@0.5: 51.8 -> 53.3 | Improved |
| add relation loss L_R alone (ablation) | SPDP-Net · JRDB-PAR | Social group detection IoU@0.5: 51.8 -> 51.4 | No clear effect |
| combine L_aux and L_R (ablation) | SPDP-Net · JRDB-PAR | Social group detection IoU@0.5: 53.3 -> 56.4 | Improved |
| increase relation loss weight to 3 (ablation) | SPDP-Net · JRDB-PAR | Overall F1: 46.5 -> 44.9 | Hurt |
| increase individual or auxiliary loss weight (ablation) | SPDP-Net · JRDB-PAR | Social activity F1: F_p 34.2 -> 29.8 (3:1:1) | Mixed |
Closest papers by meaning
- Panoramic Human Activity Recognition (ECCV 2022 · similarity 0.84)
- Decomposing and Fusing Intra- and Inter-Sensor Spatio-Temporal Signal for Multi-Sensor Wearable Human Activity Recognition (AAAI 2025 · similarity 0.79)
- When a Window Is Not an Action: Selective Phase-Script Deliberation for Sliding-Window Human Activity Recognition (NeurIPS 2026 · similarity 0.79)
- Complex Activity Recognition Via Attribute Dynamics (IJCV 2016 · similarity 0.78)
- Bi-Causal: Group Activity Recognition via Bidirectional Causality (CVPR 2024 · similarity 0.77)
- HARDVS: Revisiting Human Activity Recognition with Dynamic Vision Sensors (AAAI 2024 · similarity 0.77)
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.
