Image-based 3D target detection method and system, medium and program product

The invention discloses an image-based 3D target detection method and system, a medium and a program product, and the method comprises the steps: obtaining an image containing annotation information, and extracting the training information of the image containing annotation information, the training...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: REN WEIPING, ZHANG RUI, MEI JINREN, ZHOU YUCHEN, MENG CHAO, BIE XIAOFANG, GU KAI, WEI XI, WANG MENGJIE, ZHANG SONG
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 REN WEIPING
ZHANG RUI
MEI JINREN
ZHOU YUCHEN
MENG CHAO
BIE XIAOFANG
GU KAI
WEI XI
WANG MENGJIE
ZHANG SONG
description The invention discloses an image-based 3D target detection method and system, a medium and a program product, and the method comprises the steps: obtaining an image containing annotation information, and extracting the training information of the image containing annotation information, the training information at least comprising category information and 2.5 D target frame attribute information; obtaining a first deep learning network model and a corresponding pre-training weight, wherein the first deep learning network model at least can obtain position information and category information of the 2D target frame; generating a second deep learning network model on the first deep learning network model based on the 2.5 D target frame attribute information; loading the pre-training weight as an initial weight of the second deep learning network model, and training the second deep learning network model to obtain a 2.5 D target frame; and converting the 2.5 D target frame into a 3D target frame for 3D target de
format Patent
fullrecord <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN118115966A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN118115966A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN118115966A3</originalsourceid><addsrcrecordid>eNrjZPD1zE1MT9VNSixOTVEwdlEoSSxKTy1RSEktSU0uyczPU8hNLcnIT1FIzEtRKK4sLknN1QEKpWSW5oKFCory04sSc0F0SmlyCQ8Da1piTnEqL5TmZlB0cw1x9tBNLciPTy0uSExOzUstiXf2MzS0MDQ0tTQzczQmRg0AZ701fg</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Image-based 3D target detection method and system, medium and program product</title><source>esp@cenet</source><creator>REN WEIPING ; ZHANG RUI ; MEI JINREN ; ZHOU YUCHEN ; MENG CHAO ; BIE XIAOFANG ; GU KAI ; WEI XI ; WANG MENGJIE ; ZHANG SONG</creator><creatorcontrib>REN WEIPING ; ZHANG RUI ; MEI JINREN ; ZHOU YUCHEN ; MENG CHAO ; BIE XIAOFANG ; GU KAI ; WEI XI ; WANG MENGJIE ; ZHANG SONG</creatorcontrib><description>The invention discloses an image-based 3D target detection method and system, a medium and a program product, and the method comprises the steps: obtaining an image containing annotation information, and extracting the training information of the image containing annotation information, the training information at least comprising category information and 2.5 D target frame attribute information; obtaining a first deep learning network model and a corresponding pre-training weight, wherein the first deep learning network model at least can obtain position information and category information of the 2D target frame; generating a second deep learning network model on the first deep learning network model based on the 2.5 D target frame attribute information; loading the pre-training weight as an initial weight of the second deep learning network model, and training the second deep learning network model to obtain a 2.5 D target frame; and converting the 2.5 D target frame into a 3D target frame for 3D target de</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&amp;date=20240531&amp;DB=EPODOC&amp;CC=CN&amp;NR=118115966A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25564,76547</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&amp;date=20240531&amp;DB=EPODOC&amp;CC=CN&amp;NR=118115966A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>REN WEIPING</creatorcontrib><creatorcontrib>ZHANG RUI</creatorcontrib><creatorcontrib>MEI JINREN</creatorcontrib><creatorcontrib>ZHOU YUCHEN</creatorcontrib><creatorcontrib>MENG CHAO</creatorcontrib><creatorcontrib>BIE XIAOFANG</creatorcontrib><creatorcontrib>GU KAI</creatorcontrib><creatorcontrib>WEI XI</creatorcontrib><creatorcontrib>WANG MENGJIE</creatorcontrib><creatorcontrib>ZHANG SONG</creatorcontrib><title>Image-based 3D target detection method and system, medium and program product</title><description>The invention discloses an image-based 3D target detection method and system, a medium and a program product, and the method comprises the steps: obtaining an image containing annotation information, and extracting the training information of the image containing annotation information, the training information at least comprising category information and 2.5 D target frame attribute information; obtaining a first deep learning network model and a corresponding pre-training weight, wherein the first deep learning network model at least can obtain position information and category information of the 2D target frame; generating a second deep learning network model on the first deep learning network model based on the 2.5 D target frame attribute information; loading the pre-training weight as an initial weight of the second deep learning network model, and training the second deep learning network model to obtain a 2.5 D target frame; and converting the 2.5 D target frame into a 3D target frame for 3D target de</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>eNrjZPD1zE1MT9VNSixOTVEwdlEoSSxKTy1RSEktSU0uyczPU8hNLcnIT1FIzEtRKK4sLknN1QEKpWSW5oKFCory04sSc0F0SmlyCQ8Da1piTnEqL5TmZlB0cw1x9tBNLciPTy0uSExOzUstiXf2MzS0MDQ0tTQzczQmRg0AZ701fg</recordid><startdate>20240531</startdate><enddate>20240531</enddate><creator>REN WEIPING</creator><creator>ZHANG RUI</creator><creator>MEI JINREN</creator><creator>ZHOU YUCHEN</creator><creator>MENG CHAO</creator><creator>BIE XIAOFANG</creator><creator>GU KAI</creator><creator>WEI XI</creator><creator>WANG MENGJIE</creator><creator>ZHANG SONG</creator><scope>EVB</scope></search><sort><creationdate>20240531</creationdate><title>Image-based 3D target detection method and system, medium and program product</title><author>REN WEIPING ; ZHANG RUI ; MEI JINREN ; ZHOU YUCHEN ; MENG CHAO ; BIE XIAOFANG ; GU KAI ; WEI XI ; WANG MENGJIE ; ZHANG SONG</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN118115966A3</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>REN WEIPING</creatorcontrib><creatorcontrib>ZHANG RUI</creatorcontrib><creatorcontrib>MEI JINREN</creatorcontrib><creatorcontrib>ZHOU YUCHEN</creatorcontrib><creatorcontrib>MENG CHAO</creatorcontrib><creatorcontrib>BIE XIAOFANG</creatorcontrib><creatorcontrib>GU KAI</creatorcontrib><creatorcontrib>WEI XI</creatorcontrib><creatorcontrib>WANG MENGJIE</creatorcontrib><creatorcontrib>ZHANG SONG</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>REN WEIPING</au><au>ZHANG RUI</au><au>MEI JINREN</au><au>ZHOU YUCHEN</au><au>MENG CHAO</au><au>BIE XIAOFANG</au><au>GU KAI</au><au>WEI XI</au><au>WANG MENGJIE</au><au>ZHANG SONG</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Image-based 3D target detection method and system, medium and program product</title><date>2024-05-31</date><risdate>2024</risdate><abstract>The invention discloses an image-based 3D target detection method and system, a medium and a program product, and the method comprises the steps: obtaining an image containing annotation information, and extracting the training information of the image containing annotation information, the training information at least comprising category information and 2.5 D target frame attribute information; obtaining a first deep learning network model and a corresponding pre-training weight, wherein the first deep learning network model at least can obtain position information and category information of the 2D target frame; generating a second deep learning network model on the first deep learning network model based on the 2.5 D target frame attribute information; loading the pre-training weight as an initial weight of the second deep learning network model, and training the second deep learning network model to obtain a 2.5 D target frame; and converting the 2.5 D target frame into a 3D target frame for 3D target de</abstract><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier
ispartof
issn
language chi ; eng
recordid cdi_epo_espacenet_CN118115966A
source esp@cenet
subjects CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
PHYSICS
title Image-based 3D target detection method and system, medium and program product
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-05T02%3A01%3A21IST&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=REN%20WEIPING&rft.date=2024-05-31&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN118115966A%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