PFRL: Pose-Free Reinforcement Learning for 6D Pose Estimation
This addresses the challenge of costly data annotation in computer vision for robotics and AR/VR applications, offering a weakly-supervised method that is incremental but practical.
The paper tackles the problem of 6D pose estimation from single RGB images by eliminating the need for expensive real-world 6D pose annotations, using a reinforcement learning approach with only 2D image annotations to achieve state-of-the-art performance on LINEMOD and T-LESS datasets.
6D pose estimation from a single RGB image is a challenging and vital task in computer vision. The current mainstream deep model methods resort to 2D images annotated with real-world ground-truth 6D object poses, whose collection is fairly cumbersome and expensive, even unavailable in many cases. In this work, to get rid of the burden of 6D annotations, we formulate the 6D pose refinement as a Markov Decision Process and impose on the reinforcement learning approach with only 2D image annotations as weakly-supervised 6D pose information, via a delicate reward definition and a composite reinforced optimization method for efficient and effective policy training. Experiments on LINEMOD and T-LESS datasets demonstrate that our Pose-Free approach is able to achieve state-of-the-art performance compared with the methods without using real-world ground-truth 6D pose labels.