ParaPose: Parameter and Domain Randomization Optimization for Pose Estimation using Synthetic Data

2022-03-02 · Frederik Hagelskjaer, Anders Glent Buch

Pose estimation is the task of determining the 6D position of an object in a scene. Pose estimation aid the abilities and flexibility of robotic set-ups. However, the system must be configured towards the use case to perform adequately. This configuration is time-consuming and limits the usability of pose estimation and, thereby, robotic systems. Deep learning is a method to overcome this configuration procedure by learning parameters directly from the dataset. However, obtaining this training data can also be very time-consuming. The use of synthetic training data avoids this data collection problem, but a configuration of the training procedure is necessary to overcome the domain gap problem. Additionally, the pose estimation parameters also need to be configured. This configuration is jokingly known as grad student descent as parameters are manually adjusted until satisfactory results are obtained. This paper presents a method for automatic configuration using only synthetic data. This is accomplished by learning the domain randomization during network training, and then using the domain randomization to optimize the pose estimation parameters. The developed approach shows state-of-the-art performance of 82.0 % recall on the challenging OCCLUSION dataset, outperforming all previous methods with a large margin. These results prove the validity of automatic set-up of pose estimation using purely synthetic data.

📄 PDF Abstract BibTeX arXiv:2203.00945

Code (0)

등록된 구현이 없습니다.

Tasks

Pose Estimation

Data-efficient Domain Randomization with Bayesian Optimization

2020-03-05 · Fabio Muratore, Christian Eilers, Michael Gienger, Jan Peters

When learning policies for robot control, the required real-world data is typically prohibitively expensive to acquire, so learning in simulation is a popular strategy. Unfortunately, such polices are often not transfera…

Bayesian Optimization

Policy Transfer via Kinematic Domain Randomization and Adaptation

2020-11-03 · Ioannis Exarchos, Yifeng Jiang, Wenhao Yu, C. Karen Liu

Transferring reinforcement learning policies trained in physics simulation to the real hardware remains a challenge, known as the "sim-to-real" gap. Domain randomization is a simple yet effective technique to address dyn…

Bayesian Optimization Domain Adaptation

Active Domain Randomization

2019-04-09 · Bhairav Mehta, Manfred Diaz, Florian Golemo, Christopher J. Pal 외

Domain randomization is a popular technique for improving domain transfer, often used in a zero-shot setting when the target domain is unknown or cannot easily be used for training. In this work, we empirically examine t…

Attribute

Robust Domain Randomization for Reinforcement Learning

2019-09-25 · Reda Bahi Slaoui, William R. Clements, Jakob N. Foerster, Sébastien Toth

Producing agents that can generalize to a wide range of environments is a significant challenge in reinforcement learning. One method for overcoming this issue is domain randomization, whereby at the start of each traini…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Domain Randomization via Entropy Maximization

2023-11-03 · Gabriele Tiboni, Pascal Klink, Jan Peters, Tatiana Tommasi 외

Varying dynamics parameters in simulation is a popular Domain Randomization (DR) approach for overcoming the reality gap in Reinforcement Learning (RL). Nevertheless, DR heavily hinges on the choice of the sampling distr…

Diversity Reinforcement Learning (RL)