Sample-Guided Adaptive Class Prototype for Visual Domain Adaptation
Domain adaptation aims to handle the distribution mismatch of training and testing data, which achieves dramatic progress in multi-sensor systems. Previous methods align the cross-domain distributions by some statistics, such as the means and variances. Despite their appeal, such methods often fail...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-12-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/20/24/7036 |
_version_ | 1797545331363151872 |
---|---|
author | Chao Han Xiaoyang Li Zhen Yang Deyun Zhou Yiyang Zhao Weiren Kong |
author_facet | Chao Han Xiaoyang Li Zhen Yang Deyun Zhou Yiyang Zhao Weiren Kong |
author_sort | Chao Han |
collection | DOAJ |
description | Domain adaptation aims to handle the distribution mismatch of training and testing data, which achieves dramatic progress in multi-sensor systems. Previous methods align the cross-domain distributions by some statistics, such as the means and variances. Despite their appeal, such methods often fail to model the discriminative structures existing within testing samples. In this paper, we present a sample-guided adaptive class prototype method, which consists of the no distribution matching strategy. Specifically, two adaptive measures are proposed. Firstly, the modified nearest class prototype is raised, which allows more diversity within same class, while keeping most of the class wise discrimination information. Secondly, we put forward an easy-to-hard testing scheme by taking into account the different difficulties in recognizing target samples. Easy samples are classified and selected to assist the prediction of hard samples. Extensive experiments verify the effectiveness of the proposed method. |
first_indexed | 2024-03-10T14:13:53Z |
format | Article |
id | doaj.art-af945ae0a93947a981c6c3f22e609a48 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-10T14:13:53Z |
publishDate | 2020-12-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-af945ae0a93947a981c6c3f22e609a482023-11-20T23:58:10ZengMDPI AGSensors1424-82202020-12-012024703610.3390/s20247036Sample-Guided Adaptive Class Prototype for Visual Domain AdaptationChao Han0Xiaoyang Li1Zhen Yang2Deyun Zhou3Yiyang Zhao4Weiren Kong5School of Electronics and Information, Northwestern Polytechnical University, Xi’an 710129, ChinaSchool of Electronics and Information, Northwestern Polytechnical University, Xi’an 710129, ChinaSchool of Electronics and Information, Northwestern Polytechnical University, Xi’an 710129, ChinaSchool of Electronics and Information, Northwestern Polytechnical University, Xi’an 710129, ChinaSchool of Electronics and Information, Northwestern Polytechnical University, Xi’an 710129, ChinaSchool of Electronics and Information, Northwestern Polytechnical University, Xi’an 710129, ChinaDomain adaptation aims to handle the distribution mismatch of training and testing data, which achieves dramatic progress in multi-sensor systems. Previous methods align the cross-domain distributions by some statistics, such as the means and variances. Despite their appeal, such methods often fail to model the discriminative structures existing within testing samples. In this paper, we present a sample-guided adaptive class prototype method, which consists of the no distribution matching strategy. Specifically, two adaptive measures are proposed. Firstly, the modified nearest class prototype is raised, which allows more diversity within same class, while keeping most of the class wise discrimination information. Secondly, we put forward an easy-to-hard testing scheme by taking into account the different difficulties in recognizing target samples. Easy samples are classified and selected to assist the prediction of hard samples. Extensive experiments verify the effectiveness of the proposed method.https://www.mdpi.com/1424-8220/20/24/7036domain adaptationadaptive class prototypesample selection |
spellingShingle | Chao Han Xiaoyang Li Zhen Yang Deyun Zhou Yiyang Zhao Weiren Kong Sample-Guided Adaptive Class Prototype for Visual Domain Adaptation Sensors domain adaptation adaptive class prototype sample selection |
title | Sample-Guided Adaptive Class Prototype for Visual Domain Adaptation |
title_full | Sample-Guided Adaptive Class Prototype for Visual Domain Adaptation |
title_fullStr | Sample-Guided Adaptive Class Prototype for Visual Domain Adaptation |
title_full_unstemmed | Sample-Guided Adaptive Class Prototype for Visual Domain Adaptation |
title_short | Sample-Guided Adaptive Class Prototype for Visual Domain Adaptation |
title_sort | sample guided adaptive class prototype for visual domain adaptation |
topic | domain adaptation adaptive class prototype sample selection |
url | https://www.mdpi.com/1424-8220/20/24/7036 |
work_keys_str_mv | AT chaohan sampleguidedadaptiveclassprototypeforvisualdomainadaptation AT xiaoyangli sampleguidedadaptiveclassprototypeforvisualdomainadaptation AT zhenyang sampleguidedadaptiveclassprototypeforvisualdomainadaptation AT deyunzhou sampleguidedadaptiveclassprototypeforvisualdomainadaptation AT yiyangzhao sampleguidedadaptiveclassprototypeforvisualdomainadaptation AT weirenkong sampleguidedadaptiveclassprototypeforvisualdomainadaptation |