Category-Level Global Camera Pose Estimation with Multi-Hypothesis Point Cloud Correspondences
This addresses the challenge of accurate camera pose estimation in computer vision, particularly for category-level object registration, representing a strong incremental improvement over existing methods.
The paper tackles the problem of rigid point cloud registration by proposing a method that retains multiple possible correspondences per keypoint and gradually updates them, along with a new point feature descriptor. The method outperforms state-of-the-art approaches by up to 20% in registering real-world noisy depth images to a template shape.
Correspondence search is an essential step in rigid point cloud registration algorithms. Most methods maintain a single correspondence at each step and gradually remove wrong correspondances. However, building one-to-one correspondence with hard assignments is extremely difficult, especially when matching two point clouds with many locally similar features. This paper proposes an optimization method that retains all possible correspondences for each keypoint when matching a partial point cloud to a complete point cloud. These uncertain correspondences are then gradually updated with the estimated rigid transformation by considering the matching cost. Moreover, we propose a new point feature descriptor that measures the similarity between local point cloud regions. Extensive experiments show that our method outperforms the state-of-the-art (SoTA) methods even when matching different objects within the same category. Notably, our method outperforms the SoTA methods when registering real-world noisy depth images to a template shape by up to 20% performance.