SURE: Semi-dense Uncertainty-REfined Feature Matching
This work provides a more robust and reliable feature matching solution for robotic vision problems, particularly in challenging environments with large viewpoint changes or textureless regions, which is an incremental improvement for practitioners in robotics and computer vision.
This paper addresses the problem of unreliable image correspondences in challenging scenarios by proposing SURE, a semi-dense matching framework. SURE jointly predicts correspondences and their confidence by modeling both aleatoric and epistemic uncertainties, outperforming existing state-of-the-art semi-dense matching models in both accuracy and efficiency on multiple standard benchmarks.
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.