Multi-level Similarity Learning for Low-Shot Recognition

Low-shot learning indicates the ability to recognize unseen objects based on very limited labeled training samples, which simulates human visual intelligence. According to this concept, we propose a multi-level similarity model (MLSM) to capture the deep encoded distance metric between the support a...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2019-12
Main Authors: Xv, Hongwei, Sun, Xin, Dong, Junyu, Zhang, Shu, Li, Qiong
Format: Article
Language:English
Subjects:
Computer simulation
Image classification
Learning
Object recognition
Shot
Similarity
Target recognition
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Xv, Hongwei
Sun, Xin
Dong, Junyu
Zhang, Shu
Li, Qiong
description Low-shot learning indicates the ability to recognize unseen objects based on very limited labeled training samples, which simulates human visual intelligence. According to this concept, we propose a multi-level similarity model (MLSM) to capture the deep encoded distance metric between the support and query samples. Our approach is achieved based on the fact that the image similarity learning can be decomposed into image-level, global-level, and object-level. Once the similarity function is established, MLSM will be able to classify images for unseen classes by computing the similarity scores between a limited number of labeled samples and the target images. Furthermore, we conduct 5-way experiments with both 1-shot and 5-shot setting on Caltech-UCSD datasets. It is demonstrated that the proposed model can achieve promising results compared with the existing methods in practical applications.
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2327144259</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2327144259</sourcerecordid><originalsourceid>FETCH-proquest_journals_23271442593</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mSw8C3NKcnUzUktS81RCM7MzcxJLMosqVTwSU0sysvMS1dIyy9S8Mkv1w3OyC9RCEpNzk_PyyzJzM_jYWBNS8wpTuWF0twMym6uIc4eugVF-YWlqcUl8Vn5pUV5QKl4I2Mjc0MTEyNTS2PiVAEASIs2Fw</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2327144259</pqid></control><display><type>article</type><title>Multi-level Similarity Learning for Low-Shot Recognition</title><source>Publicly Available Content Database</source><creator>Xv, Hongwei ; Sun, Xin ; Dong, Junyu ; Zhang, Shu ; Li, Qiong</creator><creatorcontrib>Xv, Hongwei ; Sun, Xin ; Dong, Junyu ; Zhang, Shu ; Li, Qiong</creatorcontrib><description>Low-shot learning indicates the ability to recognize unseen objects based on very limited labeled training samples, which simulates human visual intelligence. According to this concept, we propose a multi-level similarity model (MLSM) to capture the deep encoded distance metric between the support and query samples. Our approach is achieved based on the fact that the image similarity learning can be decomposed into image-level, global-level, and object-level. Once the similarity function is established, MLSM will be able to classify images for unseen classes by computing the similarity scores between a limited number of labeled samples and the target images. Furthermore, we conduct 5-way experiments with both 1-shot and 5-shot setting on Caltech-UCSD datasets. It is demonstrated that the proposed model can achieve promising results compared with the existing methods in practical applications.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Computer simulation ; Image classification ; Learning ; Object recognition ; Shot ; Similarity ; Target recognition</subject><ispartof>arXiv.org, 2019-12</ispartof><rights>2019. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2327144259?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>780,784,25752,37011,44589</link.rule.ids></links><search><creatorcontrib>Xv, Hongwei</creatorcontrib><creatorcontrib>Sun, Xin</creatorcontrib><creatorcontrib>Dong, Junyu</creatorcontrib><creatorcontrib>Zhang, Shu</creatorcontrib><creatorcontrib>Li, Qiong</creatorcontrib><title>Multi-level Similarity Learning for Low-Shot Recognition</title><title>arXiv.org</title><description>Low-shot learning indicates the ability to recognize unseen objects based on very limited labeled training samples, which simulates human visual intelligence. According to this concept, we propose a multi-level similarity model (MLSM) to capture the deep encoded distance metric between the support and query samples. Our approach is achieved based on the fact that the image similarity learning can be decomposed into image-level, global-level, and object-level. Once the similarity function is established, MLSM will be able to classify images for unseen classes by computing the similarity scores between a limited number of labeled samples and the target images. Furthermore, we conduct 5-way experiments with both 1-shot and 5-shot setting on Caltech-UCSD datasets. It is demonstrated that the proposed model can achieve promising results compared with the existing methods in practical applications.</description><subject>Computer simulation</subject><subject>Image classification</subject><subject>Learning</subject><subject>Object recognition</subject><subject>Shot</subject><subject>Similarity</subject><subject>Target recognition</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mSw8C3NKcnUzUktS81RCM7MzcxJLMosqVTwSU0sysvMS1dIyy9S8Mkv1w3OyC9RCEpNzk_PyyzJzM_jYWBNS8wpTuWF0twMym6uIc4eugVF-YWlqcUl8Vn5pUV5QKl4I2Mjc0MTEyNTS2PiVAEASIs2Fw</recordid><startdate>20191213</startdate><enddate>20191213</enddate><creator>Xv, Hongwei</creator><creator>Sun, Xin</creator><creator>Dong, Junyu</creator><creator>Zhang, Shu</creator><creator>Li, Qiong</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20191213</creationdate><title>Multi-level Similarity Learning for Low-Shot Recognition</title><author>Xv, Hongwei ; Sun, Xin ; Dong, Junyu ; Zhang, Shu ; Li, Qiong</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_23271442593</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Computer simulation</topic><topic>Image classification</topic><topic>Learning</topic><topic>Object recognition</topic><topic>Shot</topic><topic>Similarity</topic><topic>Target recognition</topic><toplevel>online_resources</toplevel><creatorcontrib>Xv, Hongwei</creatorcontrib><creatorcontrib>Sun, Xin</creatorcontrib><creatorcontrib>Dong, Junyu</creatorcontrib><creatorcontrib>Zhang, Shu</creatorcontrib><creatorcontrib>Li, Qiong</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>AUTh Library subscriptions: ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Xv, Hongwei</au><au>Sun, Xin</au><au>Dong, Junyu</au><au>Zhang, Shu</au><au>Li, Qiong</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Multi-level Similarity Learning for Low-Shot Recognition</atitle><jtitle>arXiv.org</jtitle><date>2019-12-13</date><risdate>2019</risdate><eissn>2331-8422</eissn><abstract>Low-shot learning indicates the ability to recognize unseen objects based on very limited labeled training samples, which simulates human visual intelligence. According to this concept, we propose a multi-level similarity model (MLSM) to capture the deep encoded distance metric between the support and query samples. Our approach is achieved based on the fact that the image similarity learning can be decomposed into image-level, global-level, and object-level. Once the similarity function is established, MLSM will be able to classify images for unseen classes by computing the similarity scores between a limited number of labeled samples and the target images. Furthermore, we conduct 5-way experiments with both 1-shot and 5-shot setting on Caltech-UCSD datasets. It is demonstrated that the proposed model can achieve promising results compared with the existing methods in practical applications.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2019-12
issn 2331-8422
language eng
recordid cdi_proquest_journals_2327144259
source Publicly Available Content Database
subjects Computer simulation
Image classification
Learning
Object recognition
Shot
Similarity
Target recognition
title Multi-level Similarity Learning for Low-Shot Recognition
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-08T06%3A55%3A12IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Multi-level%20Similarity%20Learning%20for%20Low-Shot%20Recognition&rft.jtitle=arXiv.org&rft.au=Xv,%20Hongwei&rft.date=2019-12-13&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2327144259%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_23271442593%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2327144259&rft_id=info:pmid/&rfr_iscdi=true