Optical flow prediction method and device

Abstract

The invention provides an optical flow prediction method and device, and relates to the technical field of computer vision, and the method comprises the steps: inputting two continuous to-be-predicted images into an optical flow prediction network; obtaining predicted optical flow according to the optical flow prediction network, wherein the loss of the optical flow prediction network is determined by the following steps: cutting an input image according to a preset cutting rule, and determining two frames of cut images; inputting the two frames of cut images into an optical flow prediction network to obtain a forward optical flow and a backward optical flow; performing boundary filling mapping according to the forward optical flow, the backward optical flow and the two frames of input images, and determining two frames of mapped images after mapping; and determining the loss of the optical flow prediction network according to the cut image and the mapping image. According to the invention, the boundary filling mapping is carried out on the cut image, so the pixels moving beyond the boundary of the cut image can still be replaced by the corresponding pixels of the input image after the boundary filling mapping, the correct image loss is obtained, and the accuracy of optical flow prediction is improved.

Description

technical field [0001] The present invention relates to the technical field of computer vision, in particular to an optical flow prediction and device. Background technique [0002] With the development of computer technology, the field of computer vision has also developed quite rapidly. Optical flow prediction is a fundamental task in computer vision and plays an important role in many fields such as autonomous driving, video processing, object detection, behavior analysis, etc. In recent years, due to the rise of deep learning, optical flow prediction methods based on deep learning have gradually become mainstream. Existing optical flow prediction methods based on deep learning include FlowNet, FlowNet-v2, PWC-Net, and IRR-PWCNet. Its main advantages are high accuracy, fast speed, fast deployment, and strong generalization. However, the optical flow prediction method based on deep learning has very high requirements for training data, and requires accurate labeling of ...

AI Technical Summary

Problems solved by technology

Such a method can significantly improve the accuracy of optical flow in non-occluded areas, but it cannot solve the learning problem in occluded areas, making the estimation of optical flow in occluded areas less effective.

Images