Image semantic segmentation method based on confrontation training

A semantic segmentation and image technology, which is applied in the field of computer vision, can solve the problems of ignoring the local features of the picture, and cannot realize the meaningful fusion of the global information and local information of the image, so as to avoid the process of initializing parameters, increase interpretability, and train stably Effect

Active Publication Date: 2018-03-06
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF17 Cites 40 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the most successful network structure convolutional neural network (Convolutional Neural Network, CNN) in the field of computer vision has a major disadvantage for image semantic segmentation: due to the large number of maximum pooling layers stacked in the network structure, CNN finally obtains The feature is the informat

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
  • Image semantic segmentation method based on confrontation training
  • Image semantic segmentation method based on confrontation training
  • Image semantic segmentation method based on confrontation training

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] combined with figure 1 , the image semantic segmentation method based on adversarial training of the present embodiment, comprises the following steps:

[0045] Step 1: Input the original image to the convolutional neural network (i.e., the generation network G) for forward transfer to obtain a low-resolution segmented image;

[0046] Specifically: set the size of the original image as H×W×3, input the original image to the convolutional neural network (that is, the generation network G) for convolution pooling operation, and obtain the first downsampling feature layer with a size of H / the s 1 ×W / s 1 ×C down1 , and then perform convolution and pooling operations on the first downsampled feature layer again to obtain the second downsampled feature layer with a size of H / (s 1 ×s 2 )×W / (s 1 ×s 2 )×C down2 , repeating this process, the third downsampling feature layer, the fourth downsampling feature layer, etc. can be obtained in turn. For the sake of simplicity, ...

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 belongs to the technical field of computer vision, discloses an image semantic segmentation method based on confrontation training and is used for solving a problem that the existing semantic segmentation method cannot achieve meaningful fusion of global information and local information of images and cannot perform learning on high-order potential energy in the images. According tothe invention, a loss function of the whole network is defined based on a confrontation training network to serve as a confrontation network of a general function approximator, thereby not only beingcapable of learning how to combine different levels of information, but also being capable of forcing the generated network to learn information such as single points, paring and high-order potentialenergy in a segmented picture, achieving organic integration of local features and global features of the image and acquiring a segmentation image with the effect being more authentic; and meanwhile,the layer-by-layer training method avoids the complex network parameter initialization process, and enabling the whole network to use a method of random initialization.

Description

technical field [0001] The invention belongs to the technical field of computer vision and relates to image semantic segmentation and confrontation training, in particular to an image semantic segmentation method based on confrontation training. Background technique [0002] With the development and popularization of artificial intelligence, the important position of image semantic segmentation in the field of computer vision has become increasingly prominent. Many applications require accurate and efficient segmentation techniques, such as autonomous driving, indoor navigation, human-computer interaction, and more. In the past five years, deep learning methods have achieved great success in the field of computer vision, and various network structures have been proposed to solve different problems in this field, such as image classification and positioning. However, the most successful network structure convolutional neural network (Convolutional Neural Network, CNN) in the...

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
IPC IPC(8): G06T7/11G06K9/62G06N3/04
CPCG06T7/11G06N3/045G06F18/2415G06F18/214
Inventor 高建彬邓泽露
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
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