Matching screening method, device, electronic device and computer readable storage medium

Abstract

An embodiment of the present application provides a matching and screening method, device, electronic device, and computer-readable storage medium. The matching and screening method includes: the electronic device obtains an initial matching set, and the initial matching set is derived from an initial matching result between image pairs; At least one clipping module selects a matching subset from the initial matching set, the correct matching ratio in the matching subset is higher than the correct matching ratio in the initial matching set, and at least one clipping module is used to obtain the consistency of each initial matching in the initial matching set sex information; this matching subset is used to process image tasks associated with image pairs. The embodiment of the present application can improve the processing effect of the parametric transformation model processing image task.

Description

technical field

[0001] The present application relates to the field of image processing, in particular to a matching screening method, device, electronic equipment and computer-readable storage medium. Background technique

[0002] In the field of computer vision and image processing, feature matching is one of the basic research problems. The matching in the initial matching set is generally based on the Euclidean distance similarity between the descriptors corresponding to the matching points between image pairs. From the two groups of image pairs Select matching points among the feature points, and this matching method often has a large number of false matches.

[0003] At present, the neural network model of deep learning is generally learned and trained based on the initial matching set and the corresponding image tasks are performed. Since the distribution of samples in the initial matching set is often unbalanced, if the number of wrong matches in the initial matchin...

Images