Luxury appraisal model training method, luxury appraisal method and device
The invention provides a luxury authentication model training method, a luxury authentication method and a luxury authentication device, and the model training method comprises the steps: obtaining a main task data set, and generating an auxiliary task data set based on the main task data set; the m...
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Patent |
| Sprache: | chi ; eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | |
|---|---|
| container_issue | |
| container_start_page | |
| container_title | |
| container_volume | |
| creator | LI ZHIHAO YANG YU YANG YAOTIAN GUO XINSHENG LEI MIN CHEN XIUBO ZOU JIXIN ZHOU LINNA |
| description | The invention provides a luxury authentication model training method, a luxury authentication method and a luxury authentication device, and the model training method comprises the steps: obtaining a main task data set, and generating an auxiliary task data set based on the main task data set; the main task data set and the auxiliary task data set form a luxury character region image and a training set corresponding to two types of tags, one type of tags identifies whether the characters are true or false, and the other type of tags identifies whether the character forms change or not; the luxury identification model is trained based on the training set, a trained luxury identification model is obtained, the model adopts a multi-task learning mechanism comprising a main branch and an auxiliary branch, the main task branch is used for identifying authenticity of characters, and the auxiliary task branch is used for identifying whether character forms change or not; the two branches share the same feature extra |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN117392684A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN117392684A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN117392684A3</originalsourceid><addsrcrecordid>eNrjZPDyKa0oLapUSCwoKErMLE7MUcjNT0nNUSgB8vIy89IVclNLMvJTdBRyMNSBJRQS81IUUlLLMpNTeRhY0xJzilN5oTQ3g6Kba4izh25qQX58anFBYnJqXmpJvLOfoaG5saWRmYWJozExagBUUDVt</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Luxury appraisal model training method, luxury appraisal method and device</title><source>esp@cenet</source><creator>LI ZHIHAO ; YANG YU ; YANG YAOTIAN ; GUO XINSHENG ; LEI MIN ; CHEN XIUBO ; ZOU JIXIN ; ZHOU LINNA</creator><creatorcontrib>LI ZHIHAO ; YANG YU ; YANG YAOTIAN ; GUO XINSHENG ; LEI MIN ; CHEN XIUBO ; ZOU JIXIN ; ZHOU LINNA</creatorcontrib><description>The invention provides a luxury authentication model training method, a luxury authentication method and a luxury authentication device, and the model training method comprises the steps: obtaining a main task data set, and generating an auxiliary task data set based on the main task data set; the main task data set and the auxiliary task data set form a luxury character region image and a training set corresponding to two types of tags, one type of tags identifies whether the characters are true or false, and the other type of tags identifies whether the character forms change or not; the luxury identification model is trained based on the training set, a trained luxury identification model is obtained, the model adopts a multi-task learning mechanism comprising a main branch and an auxiliary branch, the main task branch is used for identifying authenticity of characters, and the auxiliary task branch is used for identifying whether character forms change or not; the two branches share the same feature extra</description><language>chi ; eng</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; PHYSICS</subject><creationdate>2024</creationdate><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://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20240112&DB=EPODOC&CC=CN&NR=117392684A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,309,782,887,25571,76555</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20240112&DB=EPODOC&CC=CN&NR=117392684A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>LI ZHIHAO</creatorcontrib><creatorcontrib>YANG YU</creatorcontrib><creatorcontrib>YANG YAOTIAN</creatorcontrib><creatorcontrib>GUO XINSHENG</creatorcontrib><creatorcontrib>LEI MIN</creatorcontrib><creatorcontrib>CHEN XIUBO</creatorcontrib><creatorcontrib>ZOU JIXIN</creatorcontrib><creatorcontrib>ZHOU LINNA</creatorcontrib><title>Luxury appraisal model training method, luxury appraisal method and device</title><description>The invention provides a luxury authentication model training method, a luxury authentication method and a luxury authentication device, and the model training method comprises the steps: obtaining a main task data set, and generating an auxiliary task data set based on the main task data set; the main task data set and the auxiliary task data set form a luxury character region image and a training set corresponding to two types of tags, one type of tags identifies whether the characters are true or false, and the other type of tags identifies whether the character forms change or not; the luxury identification model is trained based on the training set, a trained luxury identification model is obtained, the model adopts a multi-task learning mechanism comprising a main branch and an auxiliary branch, the main task branch is used for identifying authenticity of characters, and the auxiliary task branch is used for identifying whether character forms change or not; the two branches share the same feature extra</description><subject>CALCULATING</subject><subject>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</subject><subject>COMPUTING</subject><subject>COUNTING</subject><subject>PHYSICS</subject><fulltext>true</fulltext><rsrctype>patent</rsrctype><creationdate>2024</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNrjZPDyKa0oLapUSCwoKErMLE7MUcjNT0nNUSgB8vIy89IVclNLMvJTdBRyMNSBJRQS81IUUlLLMpNTeRhY0xJzilN5oTQ3g6Kba4izh25qQX58anFBYnJqXmpJvLOfoaG5saWRmYWJozExagBUUDVt</recordid><startdate>20240112</startdate><enddate>20240112</enddate><creator>LI ZHIHAO</creator><creator>YANG YU</creator><creator>YANG YAOTIAN</creator><creator>GUO XINSHENG</creator><creator>LEI MIN</creator><creator>CHEN XIUBO</creator><creator>ZOU JIXIN</creator><creator>ZHOU LINNA</creator><scope>EVB</scope></search><sort><creationdate>20240112</creationdate><title>Luxury appraisal model training method, luxury appraisal method and device</title><author>LI ZHIHAO ; YANG YU ; YANG YAOTIAN ; GUO XINSHENG ; LEI MIN ; CHEN XIUBO ; ZOU JIXIN ; ZHOU LINNA</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN117392684A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>chi ; eng</language><creationdate>2024</creationdate><topic>CALCULATING</topic><topic>COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS</topic><topic>COMPUTING</topic><topic>COUNTING</topic><topic>PHYSICS</topic><toplevel>online_resources</toplevel><creatorcontrib>LI ZHIHAO</creatorcontrib><creatorcontrib>YANG YU</creatorcontrib><creatorcontrib>YANG YAOTIAN</creatorcontrib><creatorcontrib>GUO XINSHENG</creatorcontrib><creatorcontrib>LEI MIN</creatorcontrib><creatorcontrib>CHEN XIUBO</creatorcontrib><creatorcontrib>ZOU JIXIN</creatorcontrib><creatorcontrib>ZHOU LINNA</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>LI ZHIHAO</au><au>YANG YU</au><au>YANG YAOTIAN</au><au>GUO XINSHENG</au><au>LEI MIN</au><au>CHEN XIUBO</au><au>ZOU JIXIN</au><au>ZHOU LINNA</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Luxury appraisal model training method, luxury appraisal method and device</title><date>2024-01-12</date><risdate>2024</risdate><abstract>The invention provides a luxury authentication model training method, a luxury authentication method and a luxury authentication device, and the model training method comprises the steps: obtaining a main task data set, and generating an auxiliary task data set based on the main task data set; the main task data set and the auxiliary task data set form a luxury character region image and a training set corresponding to two types of tags, one type of tags identifies whether the characters are true or false, and the other type of tags identifies whether the character forms change or not; the luxury identification model is trained based on the training set, a trained luxury identification model is obtained, the model adopts a multi-task learning mechanism comprising a main branch and an auxiliary branch, the main task branch is used for identifying authenticity of characters, and the auxiliary task branch is used for identifying whether character forms change or not; the two branches share the same feature extra</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | chi ; eng |
| recordid | cdi_epo_espacenet_CN117392684A |
| source | esp@cenet |
| subjects | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING PHYSICS |
| title | Luxury appraisal model training method, luxury appraisal method and device |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-04T05%3A11%3A11IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-epo_EVB&rft_val_fmt=info:ofi/fmt:kev:mtx:patent&rft.genre=patent&rft.au=LI%20ZHIHAO&rft.date=2024-01-12&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN117392684A%3C/epo_EVB%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |