Search papers, labs, and topics across Lattice.
This paper introduces TriMatch, a multi-source feature fusion framework designed to enhance two-view correspondence learning by addressing the challenges posed by pseudo-consistent outliers in complex scenes. By jointly extracting and aligning geometric, texture semantic, and structural semantic features, TriMatch improves the discrimination of true correspondences from false ones. Experimental results show that TriMatch significantly outperforms existing methods in terms of effectiveness, robustness, and generalization across various challenging scenarios.
TriMatch achieves superior correspondence discrimination by integrating geometric and semantic features, effectively tackling the issue of pseudo-consistent outliers in complex scenes.
Two-view correspondence learning aims to distinguish true correspondences (inliers) from false ones (outliers) in image pairs by leveraging their underlying differences. Existing methods mainly rely on coordinate-based geometric consistency. However, they often struggle with pseudo-consistent outliers in scenes containing repetitive structures, textureless regions, or locally similar geometric patterns. To address this limitation, we propose TriMatch, a multi-source feature fusion framework for two-view correspondence learning, which consists of two parts: feature extraction and feature refinement. In feature extraction, TriMatch jointly extracts geometric, texture semantic, and structural semantic features to provide complementary evidence for correspondence discrimination. To bridge the gap between semantic and geometric features, texture and structural semantic features are aligned with geometric features through dedicated Texture-Geometric Alignment and Structural-Geometric Alignment modules, respectively. We further introduce a Semantic-Guided Correspondence Modulation module, which modulates geometric features using semantic information to suppress geometrically plausible but semantically inconsistent correspondences. In feature refinement, a Hierarchical Semantic-Enhanced Correspondence Refinement strategy progressively models correspondence dependencies and recalibrates multi-context feature responses, enabling more reliable inlier-outlier discrimination. Extensive experiments demonstrate the effectiveness, robustness, and generalization capability of TriMatch.
