# Active 6D Multi-Object Pose Estimation in Cluttered Scenarios with Deep Reinforcement Learning

arXiv: https://arxiv.org/abs/1910.08811 · HTML: https://mlsanity.com/paper/1910.08811/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Oct 2019.

## Built on by (most cited first, 6 of 6)

- [Adaptive Rotated Convolution for Rotated Object Detection](https://mlsanity.com/paper/2303.07820/) (2023)
- [PFRL: Pose-Free Reinforcement Learning for 6D Pose Estimation](https://mlsanity.com/paper/2102.12096/) (2021)
- [Glimpse-Attend-and-Explore: Self-Attention for Active Visual Exploration](https://mlsanity.com/paper/2108.11717/) (2021)
- [Next-Best-View Prediction for Active Stereo Cameras and Highly Reflective Objects](https://mlsanity.com/paper/2202.13263/) (2022)
- [Active 6D Pose Estimation for Textureless Objects using Multi-View RGB Frames](https://mlsanity.com/paper/2503.03726/) (2025)
- [Active Classification of Moving Targets with Learned Control Policies](https://mlsanity.com/paper/2212.03068/) (2022)

## Abstract (arXiv)

In this work, we explore how a strategic selection of camera movements can facilitate the task of 6D multi-object pose estimation in cluttered scenarios while respecting real-world constraints important in robotics and augmented reality applications, such as time and distance traveled. In the proposed framework, a set of multiple object hypotheses is given to an agent, which is inferred by an object pose estimator and subsequently spatio-temporally selected by a fusion function that makes use of a verification score that circumvents the need of ground-truth annotations. The agent reasons about these hypotheses, directing its attention to the object which it is most uncertain about, moving the camera towards such an object. Unlike previous works that propose short-sighted policies, our agent is trained in simulated scenarios using reinforcement learning, attempting to learn the camera moves that produce the most accurate object poses hypotheses for a given temporal and spatial budget, without the need of viewpoints rendering during inference. Our experiments show that the proposed approach successfully estimates the 6D object pose of a stack of objects in both challenging cluttered synthetic and real scenarios, showing superior performance compared to strong baselines.