# Towards Accurate Human Pose Estimation in Videos of Crowded Scenes

arXiv: https://arxiv.org/abs/2010.10008 · HTML: https://mlsanity.com/paper/2010.10008/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Oct 2020.

## Built on by (most cited first, 7 of 7)

- [2D Human Pose Estimation: A Survey](https://mlsanity.com/paper/2204.07370/) (2022)
- [Mutual Information-Based Temporal Difference Learning for Human Pose Estimation in Video](https://mlsanity.com/paper/2303.08475/) (2023)
- [Large Scale Real-World Multi-Person Tracking](https://mlsanity.com/paper/2211.02175/) (2022)
- [SHaRPose: Sparse High-Resolution Representation for Human Pose Estimation](https://mlsanity.com/paper/2312.10758/) (2023)
- [SMART-Vision: Survey of Modern Action Recognition Techniques in Vision](https://mlsanity.com/paper/2501.13066/) (2025)
- [Relation-Based Associative Joint Location for Human Pose Estimation in Videos](https://mlsanity.com/paper/2107.03591/) (2021)
- [Semantic and Articulated Pedestrian Sensing Onboard a Moving Vehicle](https://mlsanity.com/paper/2309.06313/) (2023)

## Abstract (arXiv)

Video-based human pose estimation in crowded scenes is a challenging problem due to occlusion, motion blur, scale variation and viewpoint change, etc. Prior approaches always fail to deal with this problem because of (1) lacking of usage of temporal information; (2) lacking of training data in crowded scenes. In this paper, we focus on improving human pose estimation in videos of crowded scenes from the perspectives of exploiting temporal context and collecting new data. In particular, we first follow the top-down strategy to detect persons and perform single-person pose estimation for each frame. Then, we refine the frame-based pose estimation with temporal contexts deriving from the optical-flow. Specifically, for one frame, we forward the historical poses from the previous frames and backward the future poses from the subsequent frames to current frame, leading to stable and accurate human pose estimation in videos. In addition, we mine new data of similar scenes to HIE dataset from the Internet for improving the diversity of training set. In this way, our model achieves best performance on 7 out of 13 videos and 56.33 average wAP on test dataset of HIE challenge.