Unlock instant, AI-driven research and patent intelligence for your innovation.

Monocular depth estimation network optimization method based on contrast learning

A technology of depth estimation and optimization method, applied in the field of computer vision, which can solve the problems of limited single-frame image features and inability to accurately restore the absolute scale.

Pending Publication Date: 2021-07-06
DUT ARTIFICIAL INTELLIGENCE INST DALIAN +1
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although a single frame image also has a lot of information that can help the network judge the depth value, it still cannot accurately restore the absolute scale, and the features that a single frame image can provide are very limited.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Monocular depth estimation network optimization method based on contrast learning
  • Monocular depth estimation network optimization method based on contrast learning
  • Monocular depth estimation network optimization method based on contrast learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, other embodiments obtained by persons of ordinary skill in the art without making creative efforts all belong to the protection scope of the present invention.

[0032] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "vertical", "upper", "lower", "horizontal" etc. is based on the orientation or positional relationship shown in the drawings, and is only In order to facilitate the description of the present invention and simplify the description, it does not indicate or imply that the device or element referred to must have a specific o...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a monocular depth estimation network optimization method based on contrast learning. The method comprises the following steps: organizing a data set; predicting by adopting an unsupervised depth estimation network to obtain an initial depth map; performing data preprocessing on the initial depth map; optimizing the initial depth map by using a time sequence reference network; and calculating a loss function, designing a consistency loss function according to the idea of contrast learning, and optimizing the whole network. In the aspect of network training, a consistency loss function is designed by referring to the training thought of contrast learning, and the output of different information sources is subjected to consistency constraint, so that the feature expression of the network is increased, and the accuracy of depth estimation is improved.

Description

technical field [0001] The invention relates to the field of computer vision based on convolutional neural networks, in particular to an optimization method for a monocular depth estimation network based on contrastive learning. Background technique [0002] In recent years, with the maturity of deep learning related theories and the popularization of mobile camera equipment, related applications in the field of computer vision have made rapid progress. At the same time, people's demands on the visual field are also getting higher and higher. People are not satisfied with capturing the scene in the two-dimensional image, but hope to have a more vivid understanding of the three-dimensional scene expressed by the image. The depth estimation task is a very important basic task in the field of stereo vision. It aims to restore the distance information lost in the process of two-dimensional image imaging. Not only that, the accuracy of the distance information will directly aff...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/55
CPCG06T7/55G06T2207/10028G06T2207/20081Y02T10/40
Inventor 张敏李建华卢湖川
Owner DUT ARTIFICIAL INTELLIGENCE INST DALIAN