# Deep Reinforcement Learning for Active Human Pose Estimation

arXiv: https://arxiv.org/abs/2001.02024 · HTML: https://mlsanity.com/paper/2001.02024/

## Key facts

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

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

- [H3WB: Human3.6M 3D WholeBody Dataset and Benchmark](https://mlsanity.com/paper/2211.15692/) (2022)
- [Proactive Multi-Camera Collaboration For 3D Human Pose Estimation](https://mlsanity.com/paper/2303.03767/) (2023)
- [Glimpse-Attend-and-Explore: Self-Attention for Active Visual Exploration](https://mlsanity.com/paper/2108.11717/) (2021)
- [Efficient Virtual View Selection for 3D Hand Pose Estimation](https://mlsanity.com/paper/2203.15458/) (2022)
- [Active Transfer Learning for Efficient Video-Specific Human Pose Estimation](https://mlsanity.com/paper/2311.05041/) (2023)
- [OA-NBV: Occlusion-Aware Next-Best-View Planning for Human-Centered Active Perception on Mobile Robots](https://mlsanity.com/paper/2603.11072/) (2026)
- [Decanus to Legatus: Synthetic training for 2D-3D human pose lifting](https://mlsanity.com/paper/2210.02231/) (2022)

## Abstract (arXiv)

Most 3d human pose estimation methods assume that input -- be it images of a scene collected from one or several viewpoints, or from a video -- is given. Consequently, they focus on estimates leveraging prior knowledge and measurement by fusing information spatially and/or temporally, whenever available. In this paper we address the problem of an active observer with freedom to move and explore the scene spatially -- in `time-freeze' mode -- and/or temporally, by selecting informative viewpoints that improve its estimation accuracy. Towards this end, we introduce Pose-DRL, a fully trainable deep reinforcement learning-based active pose estimation architecture which learns to select appropriate views, in space and time, to feed an underlying monocular pose estimator. We evaluate our model using single- and multi-target estimators with strong result in both settings. Our system further learns automatic stopping conditions in time and transition functions to the next temporal processing step in videos. In extensive experiments with the Panoptic multi-view setup, and for complex scenes containing multiple people, we show that our model learns to select viewpoints that yield significantly more accurate pose estimates compared to strong multi-view baselines.