In this paper, we introduce a novel unsupervised domain adaptation technique for the task of 3D keypoint prediction from a single depth scan/image. Our key idea is to utilize the fact that predictions from different views of the same or similar objects should be consistent with each other. Such view consistency provides effective regularization for keypoint prediction on unlabeled instances. In addition, we introduce a geometric alignment term to regularize predictions in the target domain. The resulting loss function can be effectively optimized via alternating minimization. We demonstrate the effectiveness of our approach on real datasets and present experimental results showing that our approach is superior to state-of-the-art general-purpose domain adaptation techniques.
Unsupervised Domain Adaptation for 3D Keypoint Estimation via View Consistency
Resources: Paper »
@inproceedings{Zhou18ECCVb,
title={Unsupervised Domain Adaptation for 3D Keypoint Estimation via View Consistency},
author={Xingyi Zhou and Arjun Karpur and Chuang Gan and Linjie Luo and Qixing Huang},
booktitle={European Conference on Computer Vision (ECCV)},
year={2018}
}