A monocular depth estimation method, device, terminal and storage medium

A technology of depth estimation and target depth, applied in the field of computer vision, it can solve problems such as blurred depth map, and achieve the effect of solving edge blurring, sharpening edge boundary, and edge boundary more

Active Publication Date: 2020-11-24
HISCENE INFORMATION TECH CO LTD
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Embodiments of the present invention provide a monocular depth estimation method, device, terminal and storage medium to solve the problem that the depth map predicted by the existing monocular depth estimation network tends to be smooth and fuzzy at the edge of the depth boundary, and improve the Prediction Accuracy of Depth Maps

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
  • A monocular depth estimation method, device, terminal and storage medium
  • A monocular depth estimation method, device, terminal and storage medium
  • A monocular depth estimation method, device, terminal and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] figure 2 It is a flow chart of a monocular depth estimation method provided by Embodiment 1 of the present invention. This embodiment is applicable to the case of obtaining a high-precision depth map corresponding to a monocular image, especially for smartphones, drones, Scenarios for depth estimation from monocular images in robotics, autonomous driving technology, or augmented reality. The method can be performed by a monocular depth estimation device, which can be implemented by software and / or hardware, and integrated in a terminal that needs to estimate depth, such as drones, robots, smart phones, and the like. The method specifically includes the following steps:

[0031] S110. Acquire a monocular image to be depth estimated.

[0032] Wherein, the monocular image may refer to an image captured by a common camera. Exemplarily, the monocular image may be an RGB color image captured by an RGB (Red Green Blue) camera.

[0033] S120. Use the monocular image as the...

Embodiment 2

[0075] Figure 7 It is a schematic structural diagram of a monocular depth estimation device provided by Embodiment 2 of the present invention. This embodiment is applicable to the case of obtaining a high-precision depth map corresponding to a monocular image. The device specifically includes: a monocular image acquisition module 210 and Target depth map determination module 220;

[0076] Wherein, the monocular image acquisition module 210 is used to acquire the monocular image to be estimated in depth;

[0077] The target depth map determination module 220 is configured to use the monocular image as the input of the target depth generation model in the target generation confrontation network, and determine the target depth map corresponding to the monocular image according to the output of the target depth generation model, wherein the target depth generation The model is trained according to the deep discriminative model in the target generative adversarial network.

[00...

Embodiment 3

[0119] Figure 8 It is a schematic structural diagram of a terminal provided in Embodiment 3 of the present invention. see Figure 8 , the terminal includes:

[0120] one or more processors 310;

[0121] memory 320, for storing one or more programs;

[0122] The input device 330 is used to collect monocular images;

[0123] an output device 340, configured to display a target depth map;

[0124] When one or more programs are executed by one or more processors 310, so that one or more processors 310 implement the monocular depth estimation method provided by the embodiment of the present invention, including:

[0125] Obtain the monocular image to be estimated in depth;

[0126] The monocular image is used as the input of the target depth generation model in the target generation confrontation network, and the target depth map corresponding to the monocular image is determined according to the output of the target depth generation model, wherein the target depth generatio...

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 embodiment of the invention discloses a monocular depth estimation method, device, terminal and storage medium. The method includes: obtaining a monocular image to be estimated in depth; using the monocular image as an input of a target depth generation model in a target generation confrontation network, and determining a target depth map corresponding to the monocular image according to an output of the target depth generation model, wherein , the target deep generative model is trained according to the deep discriminative model in the target generative adversarial network. The technical solution of the embodiment of the present invention can solve the problem that the depth map predicted by the existing monocular depth estimation network tends to be smooth and fuzzy at the edge of the depth boundary, thereby improving the prediction accuracy of the depth map.

Description

technical field [0001] Embodiments of the present invention relate to computer vision technology, and in particular to a monocular depth estimation method, device, terminal and storage medium. Background technique [0002] In the field of computer vision research, more and more people are studying the monocular depth estimation method, which is to predict the distance between each position in the image and the camera through a color image acquired by a common camera (such as an RGB camera), namely depth information. [0003] With the continuous advancement of deep learning technology, more and more people use methods based on convolutional neural networks for monocular depth estimation. Through the monocular depth estimation network, the corresponding depth image can be directly obtained according to a monocular image, without using larger sensors and other equipment, which expands the scope of application. Existing monocular depth estimation networks are often trained usi...

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 Patents(China)
IPC IPC(8): G06T7/50G06N3/04
CPCG06T7/50G06T2207/20081G06T2207/20084G06T2207/10028G06N3/045
Inventor 不公告发明人
Owner HISCENE INFORMATION TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap