SURE: Semi-dense Uncertainty-REfined Feature Matching

2026-03-05

AI Summary

This work addresses the challenge that existing image matching methods often produce highly confident yet erroneous correspondences under challenging conditions such as large viewpoint changes or textureless regions, primarily due to the lack of explicit modeling of matching reliability. To this end, we propose SURE, a framework that jointly predicts image correspondences and their associated confidence by explicitly modeling both aleatoric and epistemic uncertainties. SURE introduces an evidential regression head for trustworthy coordinate prediction and incorporates a lightweight spatial fusion module to enhance local feature accuracy. This design improves matching robustness while maintaining computational efficiency. Extensive experiments demonstrate that SURE significantly outperforms state-of-the-art semi-dense matching methods across multiple standard benchmarks, achieving leading performance in both accuracy and computational efficiency.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Motion & Tracking
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Establishing reliable image correspondences is essential for many robotic vision problems. However, existing methods often struggle in challenging scenarios with large viewpoint changes or textureless regions, where incorrect cor- respondences may still receive high similarity scores. This is mainly because conventional models rely solely on fea- ture similarity, lacking an explicit mechanism to estimate the reliability of predicted matches, leading to overconfident errors. To address this issue, we propose SURE, a Semi- dense Uncertainty-REfined matching framework that jointly predicts correspondences and their confidence by modeling both aleatoric and epistemic uncertainties. Our approach in- troduces a novel evidential head for trustworthy coordinate regression, along with a lightweight spatial fusion module that enhances local feature precision with minimal overhead. We evaluated our method on multiple standard benchmarks, where it consistently outperforms existing state-of-the-art semi-dense matching models in both accuracy and efficiency. our code will be available on https://github.com/LSC-ALAN/SURE.

Problem

• feature matching
• uncertainty estimation
• image correspondence
• robust vision
• reliability assessment

Innovation

• uncertainty estimation
• feature matching
• evidential regression
• semi-dense correspondence
• spatial fusion