Learning Feature Descriptors using Camera Pose Supervision

Qianqian Wang1,2, Xiaowei Zhou3, Bharath Hariharan1, Noah Snavely1,2
1 Cornell University, 2 Cornell Tech, 3 Zhejiang University

ECCV 2020 (Oral)

[Summary video]     [Abstract]     [Paper/Supp]     [Code]     [Citation]    

Summary video


Recent research on learned visual descriptors has shown promising improvements in correspondence estimation, a key component of many 3D vision tasks. However, existing descriptor learning frameworks typically require ground-truth correspondences between feature points for training, which are challenging to acquire at scale. In this paper we propose a novel weakly-supervised framework that can learn feature descriptors solely from relative camera poses between images. To do so, we devise both a new loss function that exploits the epipolar constraint given by camera poses, and a new model architecture that makes the whole pipeline differentiable and efficient. Because we no longer need pixel-level ground-truth correspondences, our framework opens up the possibility of training on much larger and more diverse datasets for better and unbiased descriptors. We call the resulting descriptors CAmera Pose Supervised, or CAPS, descriptors. Though trained with weak supervision, CAPS descriptors outperform even prior fully-supervised descriptors and achieve state-of-the-art performance on a variety of geometric tasks.


[Supplementary material]


Please refer to our  github repo for code, pretrained model and training data.


  Title = {Learning Feature Descriptors using Camera Pose Supervision},
  Author = {Qianqian Wang and Xiaowei Zhou and Bharath Hariharan and Noah Snavely},
  booktitle = {Proc. European Conference on Computer Vision (ECCV)},
  Year = {2020},