CorrAdaptor: Adaptive Local Context Learning for Correspondence Pruning

TL;DR

CorrAdaptor introduces an adaptive local context learning architecture with dual-branch graph learning and motion injection, significantly improving correspondence pruning accuracy in computer vision tasks.

Contribution

It presents a novel dual-branch architecture with explicit and implicit graph learning, and a motion injection module, enhancing robustness and adaptability in correspondence pruning.

Findings

Achieves state-of-the-art performance on correspondence tasks.

Improves robustness to complex image variations.

Enhances outlier suppression and local context refinement.

Abstract

In the fields of computer vision and robotics, accurate pixel-level correspondences are essential for enabling advanced tasks such as structure-from-motion and simultaneous localization and mapping. Recent correspondence pruning methods usually focus on learning local consistency through k-nearest neighbors, which makes it difficult to capture robust context for each correspondence. We propose CorrAdaptor, a novel architecture that introduces a dual-branch structure capable of adaptively adjusting local contexts through both explicit and implicit local graph learning. Specifically, the explicit branch uses KNN-based graphs tailored for initial neighborhood identification, while the implicit branch leverages a learnable matrix to softly assign neighbors and adaptively expand the local context scope, significantly enhancing the model's robustness and adaptability to complex image variations. Moreover, we design a motion injection module to integrate motion consistency into the network to suppress the impact of outliers and refine local context learning, resulting in substantial performance improvements. The experimental results on extensive correspondence-based tasks indicate that our CorrAdaptor achieves state-of-the-art performance both qualitatively and quantitatively. The code and pre-trained models are available at https://github.com/TaoWangzj/CorrAdaptor.