Depth map enhancement method and device

Abstract

The invention discloses a depth map enhancement method and device, and the method comprises the steps: obtaining an initial depth map from original visual data, carrying out the alternate convolution of the initial depth map, carrying out the multi-scale feature extraction of a deconvolution module, obtaining a feature view, carrying out the scale compression and convolution, and sequentially obtaining the feature vectors of two stages; and performing spatial transformation and feature extraction on the initial depth map to obtain depth map features, reinforcing the depth map features based on the feature vectors of the two stages, recovering a depth structure, generating a feature map, and further obtaining a final depth map through multi-layer perceptron mapping. Therefore, the technical problems that in the prior art, due to the fact that the target object needs to meet certain specific geometric symmetry and then feature extraction and prediction are conducted, it is difficult to use a universal enhancement model to conduct completion and densification of geometric features in the depth collection process, and information is incomplete and precision is low in collection are solved.

Description

technical field [0001] The present application relates to the technical field of computer vision, and in particular, to a depth map enhancement method and device. Background technique [0002] With the advancement of technology, the acquisition and acquisition of stereo object models has become easier and easier, especially depth maps have attracted more and more research attention due to their applications in various fields such as autonomous driving and robotics. [0003] Due to many factors such as low resolution and occlusion of the device, the quality of depth data directly obtained by depth scanning devices such as lidar and depth cameras is usually poor. Low-quality data is mainly manifested in incomplete depth spatial structure and information redundancy, etc. Therefore, enhancement methods and systems for depth maps are becoming more and more important in practical engineering applications. [0004] In the related art, the completion and densification methods based...

AI Technical Summary

Problems solved by technology

[0005] This application provides a depth map enhancement method and device to solve the problem of difficulty in using general enhancements in the process of depth acquisition due to the fact that the target object needs to satisfy a certain geometric symmetry before performing feature extraction and prediction. The model completes and densifies the geometric features, which causes technical problems such as incomplete information and low precision in the collection

Images