CasP: Improving Semi-Dense Feature Matching Pipeline Leveraging Cascaded Correspondence Priors for Guidance

TL;DR

CasP introduces a cascaded correspondence prior-guided semi-dense feature matching pipeline that enhances accuracy and efficiency, especially at high resolutions, with superior geometric estimation and cross-domain generalization.

Contribution

The paper proposes a novel two-phase matching pipeline with region-based cross-attention and high-level features, significantly improving speed and accuracy over existing methods.

Findings

Achieves ~2.2x speedup at 1152 resolution compared to ELoFTR.

Demonstrates superior geometric estimation and cross-domain generalization.

Effective for latency-sensitive and high-robustness applications like SLAM and UAVs.

Abstract

Semi-dense feature matching methods have shown strong performance in challenging scenarios. However, the existing pipeline relies on a global search across the entire feature map to establish coarse matches, limiting further improvements in accuracy and efficiency. Motivated by this limitation, we propose a novel pipeline, CasP, which leverages cascaded correspondence priors for guidance. Specifically, the matching stage is decomposed into two progressive phases, bridged by a region-based selective cross-attention mechanism designed to enhance feature discriminability. In the second phase, one-to-one matches are determined by restricting the search range to the one-to-many prior areas identified in the first phase. Additionally, this pipeline benefits from incorporating high-level features, which helps reduce the computational costs of low-level feature extraction. The acceleration gains of CasP increase with higher resolution, and our lite model achieves a speedup of $\sim2.2\times$ at a resolution of 1152 compared to the most efficient method, ELoFTR. Furthermore, extensive experiments demonstrate its superiority in geometric estimation, particularly with impressive cross-domain generalization. These advantages highlight its potential for latency-sensitive and high-robustness applications, such as SLAM and UAV systems. Code is available at https://github.com/pq-chen/CasP.