Monocular image depth-of-field real-time calculation method based on unsupervised deep learning

Abstract

The invention discloses a monocular image depth-of-field real-time calculation method based on unsupervised deep learning, and the method comprises the steps: constructing a supervision signal throughemploying a geometric constraint relationship between binocular sequence images, replacing a conventional manual mark data set, and completing the design of an unsupervised algorithm. In a Depth-CNNnetwork, a loss function not only considers geometric constraints between images, but also designs depth-of-field estimation result consistency constraint terms for the left image and the right image,so that the algorithm accuracy is improved; the output of Depth-CNN as a part of the Pose-CNN input to construct an overall objective function, and the geometric relationship between binocular imagesand the geometric relationship between sequence images are used to construct a supervision signal, thereby further improving the accuracy and robustness of the algorithm.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a method for real-time calculation of monocular image depth of field based on unsupervised deep learning. Background technique [0002] Due to its low purchase price and real-time acquisition of complete scene information, cameras are widely used in the research of scene perception technology for service robots, autonomous navigation robots, and unmanned vehicles. With the development of high-performance computing equipment, artificial intelligence technology that uses deep neural networks to analyze 2D image information is increasingly playing an irreplaceable role in areas such as unmanned driving and robot navigation. Among them, the real-time calculation of scene depth based on monocular images is the premise of 3D scene perception technology. David Eigen first used the deep neural network to calculate the scene depth corresponding to the 2D image in 2014, an...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problem of 3D scene perception in outdoor unmanned vehicles or unmanned autonomous robots, the manual marking data set is not easy to obtain in large quantities and the application scenarios are limited, etc., the present invention provides a real-time calculation of monocular image depth of field based on unsupervised deep learning method, which only uses unlabeled images as a training data set to complete the method of accurately and quickly estimating the scene depth

Images