Robust 6D Object Pose Estimation by Learning RGB-D Features
This addresses a key challenge in robotic manipulation and grasping by improving pose estimation accuracy, though it appears incremental as it builds on existing methods to handle rotation ambiguity.
The paper tackles the problem of 6D object pose estimation, particularly for symmetric objects, by proposing a discrete-continuous formulation for rotation regression to avoid local optima, and it outperforms state-of-the-art methods on LINEMOD and YCB-Video benchmarks.
Accurate 6D object pose estimation is fundamental to robotic manipulation and grasping. Previous methods follow a local optimization approach which minimizes the distance between closest point pairs to handle the rotation ambiguity of symmetric objects. In this work, we propose a novel discrete-continuous formulation for rotation regression to resolve this local-optimum problem. We uniformly sample rotation anchors in SO(3), and predict a constrained deviation from each anchor to the target, as well as uncertainty scores for selecting the best prediction. Additionally, the object location is detected by aggregating point-wise vectors pointing to the 3D center. Experiments on two benchmarks: LINEMOD and YCB-Video, show that the proposed method outperforms state-of-the-art approaches. Our code is available at https://github.com/mentian/object-posenet.