English

Event-Based Feature Tracking in Continuous Time with Sliding Window Optimization

Computer Vision and Pattern Recognition 2021-07-12 v1

Abstract

We propose a novel method for continuous-time feature tracking in event cameras. To this end, we track features by aligning events along an estimated trajectory in space-time such that the projection on the image plane results in maximally sharp event patch images. The trajectory is parameterized by nthn^{th} order B-splines, which are continuous up to (n2)th(n-2)^{th} derivative. In contrast to previous work, we optimize the curve parameters in a sliding window fashion. On a public dataset we experimentally confirm that the proposed sliding-window B-spline optimization leads to longer and more accurate feature tracks than in previous work.

Keywords

Cite

@article{arxiv.2107.04536,
  title  = {Event-Based Feature Tracking in Continuous Time with Sliding Window Optimization},
  author = {Jason Chui and Simon Klenk and Daniel Cremers},
  journal= {arXiv preprint arXiv:2107.04536},
  year   = {2021}
}

Comments

9 pages, 4 figures, 1 table

R2 v1 2026-06-24T04:02:53.986Z