计算机科学
匹配(统计)
特征(语言学)
人工智能
模式识别(心理学)
数学
统计
语言学
哲学
作者
Johan Edstedt,Qiyu Sun,Georg Bökman,Mårten Wadenbäck,Michael Felsberg
出处
期刊:Proceedings
[Institute of Electrical and Electronics Engineers]
日期:2024-06-16
卷期号:: 19790-19800
被引量:690
标识
DOI:10.1109/cvpr52733.2024.01871
摘要
Feature matching is an important computer vision task that involves estimating correspondences between two images of a 3D scene, and dense methods estimate all such correspondences. The aim is to learn a robust model, i.e., a model able to match under challenging real-world changes. In this work, we propose such a model, leveraging frozen pretrained features from the foundation model DINOv2. Al-though these features are significantly more robust than local features trained from scratch, they are inherently coarse. We therefore combine them with specialized ConvNet fine features, creating a precisely localizable feature pyramid. To further improve robustness, we propose a tailored transformer match decoder that predicts anchor probabilities, which enables it to express multimodality. Finally, we propose an improved loss formulation through regression-by-classification with subsequent robust regression. We conduct a comprehensive set of experiments that show that our method, RoMa, achieves significant gains, setting a new state-of-the-art. In particular, we achieve a 36% improvement on the extremely challenging WxBS benchmark. Code is provided at github.com/Parskatt/RoMa.
科研通智能强力驱动
Strongly Powered by AbleSci AI