OnePose++: Keypoint-Free One-Shot Object Pose Estimation without CAD Models

• DOI: 10.48550/arxiv.2301.07673
• Publication Date: 2023-01-01
• Authors: Xingyi He; Jiaming Sun; Yuang Wang; Di Huang; Hujun Bao; Xiaowei Zhou
• Keywords: Feature (linguistics); Margin (machine learning)

We propose a new method for object pose estimation without CAD models. The previous feature-matching-based OnePose has shown promising results under one-shot setting which eliminates the need models or object-specific training. However, relies on detecting repeatable image keypoints and is thus prone to failure low-textured objects. keypoint-free pipeline remove keypoint detection. Built upon detector-free feature matching LoFTR, we devise SfM reconstruct semi-dense point-cloud model object. Given query estimation, 2D-3D network directly establishes correspondences between reconstructed first in image. Experiments show that proposed outperforms existing CAD-model-free methods by large margin comparable CAD-model-based LINEMOD even also collect dataset composed of 80 sequences 40 objects facilitate future research estimation. supplementary material, code are available project page: https://zju3dv.github.io/onepose_plus_plus/.