English

MFuseNet: Robust Depth Estimation with Learned Multiscopic Fusion

Computer Vision and Pattern Recognition 2021-08-21 v2 Artificial Intelligence Robotics

Abstract

We design a multiscopic vision system that utilizes a low-cost monocular RGB camera to acquire accurate depth estimation. Unlike multi-view stereo with images captured at unconstrained camera poses, the proposed system controls the motion of a camera to capture a sequence of images in horizontally or vertically aligned positions with the same parallax. In this system, we propose a new heuristic method and a robust learning-based method to fuse multiple cost volumes between the reference image and its surrounding images. To obtain training data, we build a synthetic dataset with multiscopic images. The experiments on the real-world Middlebury dataset and real robot demonstration show that our multiscopic vision system outperforms traditional two-frame stereo matching methods in depth estimation. Our code and dataset are available at https://sites.google.com/view/multiscopic.

Keywords

Cite

@article{arxiv.2108.02448,
  title  = {MFuseNet: Robust Depth Estimation with Learned Multiscopic Fusion},
  author = {Weihao Yuan and Rui Fan and Michael Yu Wang and Qifeng Chen},
  journal= {arXiv preprint arXiv:2108.02448},
  year   = {2021}
}

Comments

IEEE International Conference on Robotics and Automation (ICRA) + IEEE Robotics and Automation Letters (RA-L). arXiv admin note: substantial text overlap with arXiv:2001.08212

R2 v1 2026-06-24T04:51:01.423Z