Research map / Video / Action recognition
HiEve: A Large-Scale Benchmark for Human-Centric Video Analysis in Complex Events
Weiyao Lin, Huabin Liu, Shizhan Liu, Yuxi Li, Hongkai Xiong, Guo-Jun Qi, Nicu Sebe
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 action-guided domain alignment module (ADAM) to HRNet (ablation) | HRNet · HiEve pose estimation | w-AP@avg: 52.78 -> 53.10 | Improved |
| add pose refinement module (PRM) on top of ADAM (ablation) | action-guided HRNet · HiEve pose estimation | w-AP@avg: 53.10 -> 53.92 | Improved |
| use spatial refinement (SR) only instead of both refinements (ablation) | pose refinement module · HiEve pose estimation | w-AP@avg: 53.20 vs 53.92 with SR+CR | Hurt |
| use channel refinement (CR) only (ablation) | pose refinement module · HiEve pose estimation | w-AP@avg: 53.65 vs 53.92 with SR+CR | Hurt |
| predict pose-aware motion from all three stage features m1,m2,m3 instead of m1 only (ablation) | pose-aware SlowFast · HiEve action recognition | wf-mAP@avg: 12.36 -> 13.16 | Improved |
| use m2 only instead of m1 only (ablation) | pose-aware SlowFast · HiEve action recognition | wf-mAP@avg: 12.36 -> 12.58 | Improved |
| add pose-aware feature prediction to SlowFast | SlowFast · HiEve action recognition | wf-mAP@avg: 12.08 -> 13.16 | Improved |
| add action-guided pose estimation to PoseFlow tracking | PoseFlow · HiEve pose tracking | MOTA: 44.17 -> 45.36 | Improved |
| pretrain on HiEve before COCO pose estimation | HRNet · COCO keypoints | AP: 74.4 -> 74.8 | Improved |
| pretrain on HiEve before MOT20 tracking | MOTDT · MOT20 | MOTA: 46.4 -> 47.6 | Improved |
| use ground-truth boxes (oracle) for tracking (ablation) | IOUTracker · HiEve tracking | MOTA: 38.59 -> 97.70 | Improved |
| use ground-truth boxes (oracle) for pose estimation (ablation) | DHRN · HiEve pose estimation | w-AP@avg: 52.78 -> 53.34 | Improved |
Closest papers by meaning
- Moments in Time Dataset: One Million Videos for Event Understanding (TPAMI 2019 · similarity 0.88)
- AVA: A Video Dataset of Spatio-Temporally Localized Atomic Visual Actions (CVPR 2018 · similarity 0.87)
- Chaotic World: A Large and Challenging Benchmark for Human Behavior Understanding in Chaotic Events (ICCV 2023 · similarity 0.87)
- PANDA: A Gigapixel-Level Human-Centric Video Dataset (CVPR 2020 · similarity 0.87)
- VidEvent: A Large Dataset for Understanding Dynamic Evolution of Events in Videos (AAAI 2025 · similarity 0.85)
- HARDVS: Revisiting Human Activity Recognition with Dynamic Vision Sensors (AAAI 2024 · similarity 0.84)
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.
