Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Domain adaptive migration feature method and system

An adaptive, domain-based technology, applied in the field of image processing, can solve problems such as reducing the accuracy of target recognition and ignoring label consistency information

Active Publication Date: 2019-10-18
BEIJING INSTITUTE OF TECHNOLOGYGY
View PDF3 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But like JDA, because the features learned by minimizing the distance measure that characterizes the distribution difference between the source domain and the target domain have been distorted, thereby reducing the accuracy of target recognition; and the target pseudo-label information only considers the structural consistency of the target domain , ignoring the label consistency information between the source domain and the target domain

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
  • Domain adaptive migration feature method and system
  • Domain adaptive migration feature method and system
  • Domain adaptive migration feature method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0087] In the current status of transfer learning, how to effectively reduce the distribution difference between the source domain and the target domain is the main challenge of domain adaptation. Therefore, the present invention proposes a method based on domain adaptive transfer feature learning and label consistency, which aims to pass Realize two key points to achieve the purpose of effective domain adaptation:

[0088] 1) The features learned during transfer learning are not only domain invariant but also class discriminative;

[0089] 2) Effectively utilize the pseudo-labels of the target domain to serve for domain distribution alignment.

[0090] Such as figure 1 As shown, the proposed method of the present invention integrates domain-invariant feature learning of class discriminative information and object label refinement with label consistency into a general framework. transfer in figure 1 Different shapes represent different classes.

[0091] Such as image 3 s...

Embodiment 2

[0192] In transfer learning, it is common practice to extract domain-invariant features to reduce the difference between source and target domains, so that valuable information on the source domain can be effectively transferred to the target domain. However, the features learned by minimizing the distance metric (e.g., the maximum mean difference MMD) that characterizes the distribution difference between the source and target domains have been distorted, and there is a high possibility of distortion. Feature distortion may greatly lose the intrinsic category structure information transferred from the source domain to the target domain, thereby reducing the accuracy of object recognition. Therefore, the learned features should not only be domain invariant but also class discriminative.

[0193] To achieve this goal, the present invention proposes a loss term that, for each sample in the source and target domains, penalizes the farthest data pair of its class and the closest d...

Embodiment 3

[0222] Based on the same inventive concept, this embodiment also provides a domain adaptive migration feature system, including:

[0223] The initial module is used to set the initial value for the MMD matrix and the intra-class / inter-class dispersion based on the source domain and the target domain, and set the maximum number of iterations;

[0224] Discriminant transfer feature learning module, for optimizing the source classifier based on the MMD matrix and intra-class / inter-class dispersion, and determine the corresponding pseudo-label of the target data based on the optimized source classifier;

[0225] A label consistency module, configured to obtain a target classifier based on the pseudo-labels corresponding to the source domain, the target domain, and the target data, and refine the pseudo-labels corresponding to the target data based on the target classifier;

[0226] The iterative solution module is used to update the MMD matrix and the intra-class / inter-class dispe...

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 domain adaptive migration feature method and system, and the method comprises the steps: S1, setting an initial value for an MMD matrix and intra-class / inter-class dispersionbased on a source domain and a target domain, and setting the maximum number of iterations; s2, optimizing a source classifier based on the MMD matrix and the intra-class / inter-class dispersion, and determining a pseudo label corresponding to the target data based on the optimized source classifier; s3, obtaining a target classifier based on the source domain, the target domain and the pseudo tagscorresponding to the target data, and refining the pseudo tags corresponding to the target data based on the target classifier; and S4, updating the MMD matrix and the intra-class / inter-class dispersion according to the refined pseudo tags, repeatedly executing the step S2 and the step S3 until the maximum number of iterations is reached, and setting the pseudo tags corresponding to the target data as the tags corresponding to the target data. According to the method provided by the invention, the accuracy of target domain identification is improved.

Description

technical field [0001] The present invention relates to the field of image processing, in particular to a method and system for field self-adaptive migration features. Background technique [0002] At present, aligning the domain distribution and class distribution of the source domain and the target domain, and then performing target domain identification technology, is to match the class distribution between domains by introducing pseudo-labels in the target domain, and at the same time carry out domain distribution matching, by seeking two Domain distribution adaptation and class distribution adaptation of domains achieve the effect of improving object recognition performance. [0003] Taking JDA as an example, JDA predicts the target domain pseudo-label through the source classifier, and by minimizing the distance metric MMD that characterizes the distribution difference between the source domain and the target domain, and matching the domain distribution and class distr...

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): G06N20/00G06K9/62
CPCG06N20/00G06F18/241
Inventor 李爽刘驰
Owner BEIJING INSTITUTE OF TECHNOLOGYGY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com