Occ$^2$Net: Robust Image Matching Based on 3D Occupancy Estimation for Occluded Regions

TL;DR

Occ$^2$Net introduces a 3D occupancy-based approach for robust image matching that effectively handles occluded regions by modeling occlusion relations and integrating multi-view information.

Contribution

The paper presents a novel image matching framework that models occlusion relations with 3D occupancy and infers matches in occluded areas, improving robustness in occlusion scenarios.

Findings

Outperforms state-of-the-art methods on real-world datasets

Effective in scenarios with significant occlusions

Enhances multi-view consistency through 3D occupancy modeling

Abstract

Image matching is a fundamental and critical task in various visual applications, such as Simultaneous Localization and Mapping (SLAM) and image retrieval, which require accurate pose estimation. However, most existing methods ignore the occlusion relations between objects caused by camera motion and scene structure. In this paper, we propose Occ$^2$Net, a novel image matching method that models occlusion relations using 3D occupancy and infers matching points in occluded regions. Thanks to the inductive bias encoded in the Occupancy Estimation (OE) module, it greatly simplifies bootstrapping of a multi-view consistent 3D representation that can then integrate information from multiple views. Together with an Occlusion-Aware (OA) module, it incorporates attention layers and rotation alignment to enable matching between occluded and visible points. We evaluate our method on both real-world and simulated datasets and demonstrate its superior performance over state-of-the-art methods on several metrics, especially in occlusion scenarios.