Monocular image depth estimation method and system

Abstract

The present invention relates to a monocular image depth estimation method and system. The method comprises: constructing a CNN model configuration having a feature obtaining portion as a feature obtaining configuration; arranging a plurality of full connection layers according to the required depth map resolution; connecting the full connection layers to the feature obtaining configuration to form an initial estimation network model; using a training data set to train an initial estimation network model and extract the feature map of each layer of the initial estimation network model; calculating the average relative local geometric structure errors of the feature map and the corresponding depth maps, and generating a final estimation model according to the average relative local geometric structure errors; and using the training data set and combining a loss function to perform training of the final estimation model, and performing depth detection of an input image through adoption of the final estimation model after training completion. The monocular image depth estimation method and system perform training of the CNN based on the loss function according to the average relative local geometric structure errors so as to improve the precision and the definition of the regression depth image of the CNN, and the generated depth map reserves more geometric structure features of scenes.

Description

technical field [0001] The present invention relates to the technical field of 3D image depth estimation, in particular to a monocular image depth estimation method and system. Background technique [0002] At present, most computer vision technologies are proposed on the basis of two-dimensional plane images. However, the image loses part of the information of the real three-dimensional scene during the imaging process, resulting in some inherent defects in some computer vision technologies based on two-dimensional plane images. , For example: the algorithm's understanding of the scene is wrong, and the object recognition is wrong. Therefore, it is a very important technology to extract depth information from two-dimensional plane images or video sequences to predict depth images and reconstruct three-dimensional structures. If the problem of depth estimation can be solved well, it will be of great help to computer vision technology, and it will greatly promote the applica...

AI Technical Summary

Problems solved by technology

Depth estimation algorithms based on SfM technology can usually restore higher-precision scene depth information, but their disadvantages are: subject to special scenes, camera motion must exist; when there are moving objects in the scene, the accuracy of depth calculation It also has a great impact; due to the need to solve the internal and external parameters of the camera, the speed of depth estimation is relatively slow

[0008] In the existing methods of directly estimating depth images based on CNN, a part of the depth cues related to the geometric structure of the image will be lost in the CNN regression process, and the lack of good use of these depth cues leads to the accuracy and clarity of the depth images directly predicted by CNN. limited in degree

Images