Semi-supervised Deep Large-baseline Homography Estimation with Progressive Equivalence Constraint

TL;DR

This paper introduces a semi-supervised deep learning approach for large-baseline homography estimation that employs a progressive strategy and a novel loss function, achieving state-of-the-art results in challenging scenarios.

Contribution

It proposes a progressive estimation method combined with a semi-supervised loss to improve large-baseline homography accuracy without relying solely on photometric cues.

Findings

Achieves state-of-the-art performance in large-baseline scenes.

Maintains competitive accuracy in small-baseline scenes.

Introduces a new large-scale dataset for homography estimation.

Abstract

Homography estimation is erroneous in the case of large-baseline due to the low image overlay and limited receptive field. To address it, we propose a progressive estimation strategy by converting large-baseline homography into multiple intermediate ones, cumulatively multiplying these intermediate items can reconstruct the initial homography. Meanwhile, a semi-supervised homography identity loss, which consists of two components: a supervised objective and an unsupervised objective, is introduced. The first supervised loss is acting to optimize intermediate homographies, while the second unsupervised one helps to estimate a large-baseline homography without photometric losses. To validate our method, we propose a large-scale dataset that covers regular and challenging scenes. Experiments show that our method achieves state-of-the-art performance in large-baseline scenes while keeping competitive performance in small-baseline scenes. Code and dataset are available at https://github.com/megvii-research/LBHomo.