TILDE: A Temporally Invariant Learned DEtector

TL;DR

TILDE introduces a learning-based keypoint detector that remains robust under drastic weather and lighting changes, outperforming existing methods in challenging conditions while maintaining high performance on standard datasets.

Contribution

It presents a novel training approach for a keypoint detector that is invariant to severe environmental changes, with a new dataset for evaluation.

Findings

Outperforms state-of-the-art detectors in challenging weather and lighting conditions.

Maintains competitive performance on standard datasets.

Provides a publicly available dataset for testing invariance under environmental changes.

Abstract

We introduce a learning-based approach to detect repeatable keypoints under drastic imaging changes of weather and lighting conditions to which state-of-the-art keypoint detectors are surprisingly sensitive. We first identify good keypoint candidates in multiple training images taken from the same viewpoint. We then train a regressor to predict a score map whose maxima are those points so that they can be found by simple non-maximum suppression. As there are no standard datasets to test the influence of these kinds of changes, we created our own, which we will make publicly available. We will show that our method significantly outperforms the state-of-the-art methods in such challenging conditions, while still achieving state-of-the-art performance on the untrained standard Oxford dataset.