Optical flow neural network training method and device

A neural network training and neural network technology, applied in the field of optical flow neural network training methods and devices, can solve the problems of multiple computing resources, low operating efficiency and occupation of optical flow neural network

Active Publication Date: 2020-06-19
HANGZHOU HIKVISION DIGITAL TECH
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, optical flow neural networks with higher model complexity are less efficient and may take up more computing resources

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
  • Optical flow neural network training method and device
  • Optical flow neural network training method and device
  • Optical flow neural network training method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048]The following will clearly and completely describe the technical solutions in the embodiments of the application with reference to the drawings in the embodiments of the application. Apparently, the described embodiments are only some of the embodiments of the application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

[0049] The image of an object in an image frame is static, but the object itself may be in motion. Exemplarily, a camera is used to shoot a running athlete, and in the captured image frame, the position of the pixel area where the athlete's image is located in the image frame will not change with time, while the athlete's actual Therefore, it is difficult to reflect the athlete's motion state only based on the pixel area where the image of the athlete in the image frame is located. ...

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 provides an optical flow neural network training method and device. The method comprises the following steps: inputting the sample image frame and an adjacent image frame of the sample image frame into a first neural network and a second neural network, a first dense optical flow output by the first neural network and a second dense optical flow output by the secondneural network are obtained, the first neural network is an optical flow neural network trained in advance, and the second neural network is an optical flow neural network of which the model complexity is lower than that of the first neural network; and adjusting network parameters of the second neural network based on the deviation between the first dense optical flow and the second dense optical flow. According to the embodiment of the invention, because the second neural network can learn the mapping relation between the image learned by the first neural network and the dense optical flow,the performance of the second neural network can be close to the performance of the first neural network with higher model complexity.

Description

technical field [0001] The present application relates to the technical field of computer vision, in particular to an optical flow neural network training method and device. Background technique [0002] When performing computer vision tasks, dense optical flow can be used to represent the motion state of each pixel in an image frame. In the prior art, it is possible to train an optical flow neural network, input the image frame to be processed and the adjacent image frames of the image frame to be processed into the optical flow neural network, and use the output of the optical flow neural network as the dense light flow. [0003] Since the dense optical flow includes the motion state of all pixels in the target image frame, the calculation is difficult, and an optical flow neural network with a high model complexity is required to output relatively accurate dense optical flow. However, optical flow neural networks with higher model complexity run less efficiently and may...

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): G06T7/269
CPCG06T7/269G06T2207/10016G06T2207/20081G06T2207/20084
Inventor 石大虎虞抒沁谭文明
Owner HANGZHOU HIKVISION DIGITAL TECH
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