Research map / Video / Video-language understanding
Chaotic World: A Large and Challenging Benchmark for Human Behavior Understanding in Chaotic Events
Kian Eng Ong, Xun Long Ng, Yanchao Li, Wenjie Ai, Kuangyi Zhao, Si Yong Yeo, Jun Liu
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 Dynamic Knowledge Pathfinder (2 blocks per layer) to multi-task model (ablation) | IntelliCare · Chaotic World STAL | frame-mAP@IoU0.5: 14.10 -> 16.25 | Improved |
| add multi-task learning across the four tasks (ablation) | IntelliCare · Chaotic World STAL | frame-mAP@IoU0.5: 12.06 -> 14.10 | Improved |
| add Dynamic Knowledge Pathfinder (ablation) | IntelliCare · Chaotic World STEG | m-vIoU: 31.40 -> 32.60 | Improved |
| add Dynamic Knowledge Pathfinder (ablation) | IntelliCare · Chaotic World ESSL (supervised) | cIoU: 54.66 -> 57.37 | Improved |
| add multi-task learning (ablation) | IntelliCare · Chaotic World BGA (PredCls) | R@10: 13.58 -> 15.71 | Improved |
| increase blocks per dynamic layer from 2 to 3 (ablation) | IntelliCare · Chaotic World STAL | frame-mAP@IoU0.5: 16.25 -> 16.49 | No clear effect |
| increase dynamic layers from 5 to 6 (ablation) | IntelliCare · Chaotic World STAL | frame-mAP@IoU0.5: 16.25 -> 16.42 | No clear effect |
| increase dynamic layers from 3 to 5 (ablation) | IntelliCare · Chaotic World STAL | frame-mAP@IoU0.5: 15.73 -> 16.25 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TubeDETR — TubeDETR: Spatio-Temporal Video Grounding With Transformers (CVPR 2022) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Moments in Time Dataset: One Million Videos for Event Understanding (TPAMI 2019 · similarity 0.88)
- Continuous Locomotive Crowd Behavior Generation (CVPR 2025 · similarity 0.83)
- WorldModelBench: Judging Video Generation Models As World Models (NeurIPS 2025 · similarity 0.83)
- Harmony4D: A Video Dataset for In-The-Wild Close Human Interactions (NeurIPS 2024 · similarity 0.83)
- Count What You Want: Exemplar Identification and Few-Shot Counting of Human Actions in the Wild (AAAI 2024 · similarity 0.83)
- JRDB-Social: A Multifaceted Robotic Dataset for Understanding of Context and Dynamics of Human Interactions Within Social Groups (CVPR 2024 · similarity 0.83)
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.
