Depth-Aware Multi-Grid Deep Homography Estimation with Contextual Correlation

TL;DR

This paper introduces a depth-aware multi-grid deep homography estimation method that uses a contextual correlation layer to improve robustness in low-texture and low-overlap scenes, outperforming existing solutions.

Contribution

The paper proposes a novel depth-aware multi-grid homography network with a contextual correlation layer for better accuracy in complex scenes.

Findings

Outperforms state-of-the-art methods on synthetic and real datasets.

Effectively captures long-range correlations in feature maps.

Handles depth variation and parallax in homography estimation.

Abstract

Homography estimation is an important task in computer vision applications, such as image stitching, video stabilization, and camera calibration. Traditional homography estimation methods heavily depend on the quantity and distribution of feature correspondences, leading to poor robustness in low-texture scenes. The learning solutions, on the contrary, try to learn robust deep features but demonstrate unsatisfying performance in the scenes with low overlap rates. In this paper, we address these two problems simultaneously by designing a contextual correlation layer (CCL). The CCL can efficiently capture the long-range correlation within feature maps and can be flexibly used in a learning framework. In addition, considering that a single homography can not represent the complex spatial transformation in depth-varying images with parallax, we propose to predict multi-grid homography from global to local. Moreover, we equip our network with a depth perception capability, by introducing a novel depth-aware shape-preserved loss. Extensive experiments demonstrate the superiority of our method over state-of-the-art solutions in the synthetic benchmark dataset and real-world dataset. The codes and models will be available at https://github.com/nie-lang/Multi-Grid-Deep-Homography.