Driving scene binocular depth estimation method for overcoming shielding effect

Abstract

The invention discloses a driving scene binocular depth estimation method capable of overcoming a shielding effect. According to the method, an end-to-end self-supervised deep neural network model isconstructed, images of left and right views of a driving scene are input, and disparity maps corresponding to the left and right views are output. According to the method, the geometric constraint relation between the input and the output of the deep neural network model is used for training the model, a data sample with annotation information does not need to be obtained, only a binocular camerasystem needs to be used for obtaining the image pairs of the left view and the right view, the working process is greatly simplified, the economic cost is saved, and the model can train images of moretypes of scenes. The binocular estimation method designed by the invention can effectively overcome the problems of repeated pattern textures, front and back object shielding and the like in a scene,and can obtain a depth image with relatively high precision.

Description

technical field [0001] The invention relates to the fields of machine vision and automatic driving, in particular to a binocular depth estimation method for driving scenes using self-supervised deep learning technology to overcome occlusion effects. Background technique [0002] With the advancement of artificial intelligence technology, automatic driving has been extensively studied in academia and industry. As an important part of automatic driving technology, binocular depth estimation has always been a research hotspot. The binocular depth estimation is based on the binocular camera, and the left and right views are taken, and the corresponding disparity maps are obtained from the left and right views, and then the depth image is calculated according to the binocular camera parameters. [0003] The traditional binocular depth estimation adopts the method of stereo matching to find the matching corresponding points in the left and right views. However, due to the occlusi...

AI Technical Summary

Problems solved by technology

Although this self-supervised training mechanism can avoid the trouble of collecting real scene depth information, due to the existence of occlusion effects, it is impossible for the reconstructed view to be completely consistent with the original view, which will lead to a decrease in global accuracy.

A network model trained in a self-supervised manner will be much less accurate than a network model trained in a supervised manner

Images