UDA-COPE: Unsupervised Domain Adaptation for Category-level Object Pose Estimation

Learning to estimate object pose often requires ground-truth (GT) labels, such as CAD model and absolute-scale object pose, which is expensive and laborious to obtain in the real world. To tackle this problem, we propose an unsupervised domain adaptation (UDA) for category-level object pose estimation, called UDA-COPE. Inspired by recent multi-modal UDA techniques, the proposed method exploits a teacher-student self-supervised learning scheme to train a pose estimation network without using target domain pose labels. We also introduce a bidirectional filtering method between the predicted normalized object coordinate space (NOCS) map and observed point cloud, to not only make our teacher network more robust to the target domain but also to provide more reliable pseudo labels for the student network training. Extensive experimental results demonstrate the effectiveness of our proposed method both quantitatively and qualitatively. Notably, without leveraging target-domain GT labels, our proposed method achieved comparable or sometimes superior performance to existing methods that depend on the GT labels.

PDF Abstract CVPR 2022 PDF CVPR 2022 Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
6D Pose Estimation using RGBD REAL275 UDA-COPE mAP 10, 5cm 66.0 # 5
mAP 3DIou@50 82.6 # 3
mAP 5, 5cm 34.8 # 6
mAP 3DIou@75 62.5 # 4
mAP 5, 2cm 30.4 # 4
mAP 10, 2cm 56.9 # 3

Methods


No methods listed for this paper. Add relevant methods here