English

Deep Probabilistic Feature-metric Tracking

Computer Vision and Pattern Recognition 2020-11-30 v2 Robotics

Abstract

Dense image alignment from RGB-D images remains a critical issue for real-world applications, especially under challenging lighting conditions and in a wide baseline setting. In this paper, we propose a new framework to learn a pixel-wise deep feature map and a deep feature-metric uncertainty map predicted by a Convolutional Neural Network (CNN), which together formulate a deep probabilistic feature-metric residual of the two-view constraint that can be minimised using Gauss-Newton in a coarse-to-fine optimisation framework. Furthermore, our network predicts a deep initial pose for faster and more reliable convergence. The optimisation steps are differentiable and unrolled to train in an end-to-end fashion. Due to its probabilistic essence, our approach can easily couple with other residuals, where we show a combination with ICP. Experimental results demonstrate state-of-the-art performances on the TUM RGB-D dataset and the 3D rigid object tracking dataset. We further demonstrate our method's robustness and convergence qualitatively.

Keywords

Cite

@article{arxiv.2008.13504,
  title  = {Deep Probabilistic Feature-metric Tracking},
  author = {Binbin Xu and Andrew J. Davison and Stefan Leutenegger},
  journal= {arXiv preprint arXiv:2008.13504},
  year   = {2020}
}

Comments

RAL 2020. 8 pages, 9 figures, video link: https://youtu.be/6pMosl6ZAPE

R2 v1 2026-06-23T18:12:24.810Z