Sketch recognition method and device based on fusion of three-dimensional sparse convolution and two-dimensional convolution
The invention discloses a sketch recognition method and device based on fusion of three-dimensional sparse convolution and two-dimensional convolution, and the method comprises the steps: obtaining a sketch image, carrying out the feature extraction of the sketch image, and obtaining the features of...
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 | WANG JIN ZHOU YANG LU GUODONG YANG JINGRU FANG HEMING |
| description | The invention discloses a sketch recognition method and device based on fusion of three-dimensional sparse convolution and two-dimensional convolution, and the method comprises the steps: obtaining a sketch image, carrying out the feature extraction of the sketch image, and obtaining the features of the sketch image; meanwhile, sketch point feature extraction and voxelization are conducted, by extracting geometric information of all points on strokes of a sketch, the sketch points are voxelized based on the geometric information, and voxels of the points are obtained; then, carrying out feature extraction on voxels of all points of the sketch to obtain sketch voxel features; and finally, fusing the sketch image features and the sketch voxel features, and carrying out classification identification on the sketch through a classifier. The features of the sketch and all sparse points of the sketch are fully utilized, a large blank area of the sketch is removed through Sketch-SparseVoxelNet, the spatial relation o |
| format | Patent |
| fullrecord | <record><control><sourceid>epo_EVB</sourceid><recordid>TN_cdi_epo_espacenet_CN117115459A</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>CN117115459A</sourcerecordid><originalsourceid>FETCH-epo_espacenet_CN117115459A3</originalsourceid><addsrcrecordid>eNqNjLEKwjAURbs4iPoPzw_oELSIoxTFyUX3EpMbE2zzSpPWxY-3EQfdnC7cczjT7Hm-IypLHRTfvIuOPTWIljVJr0ljcAp0lQGaRmT6kAw2FG0H5No18OmSNYVWdgGk2A9c9-9SSsQH_2hffJ5NjKwDFp-dZcvD_lIec7RcYewpeMSqPAmxEaJYF9vd6h_nBWDHSMY</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>patent</recordtype></control><display><type>patent</type><title>Sketch recognition method and device based on fusion of three-dimensional sparse convolution and two-dimensional convolution</title><source>esp@cenet</source><creator>WANG JIN ; ZHOU YANG ; LU GUODONG ; YANG JINGRU ; FANG HEMING</creator><creatorcontrib>WANG JIN ; ZHOU YANG ; LU GUODONG ; YANG JINGRU ; FANG HEMING</creatorcontrib><description>The invention discloses a sketch recognition method and device based on fusion of three-dimensional sparse convolution and two-dimensional convolution, and the method comprises the steps: obtaining a sketch image, carrying out the feature extraction of the sketch image, and obtaining the features of the sketch image; meanwhile, sketch point feature extraction and voxelization are conducted, by extracting geometric information of all points on strokes of a sketch, the sketch points are voxelized based on the geometric information, and voxels of the points are obtained; then, carrying out feature extraction on voxels of all points of the sketch to obtain sketch voxel features; and finally, fusing the sketch image features and the sketch voxel features, and carrying out classification identification on the sketch through a classifier. The features of the sketch and all sparse points of the sketch are fully utilized, a large blank area of the sketch is removed through Sketch-SparseVoxelNet, the spatial relation o</description><language>chi ; eng</language><subject>CALCULATING ; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS ; COMPUTING ; COUNTING ; PHYSICS</subject><creationdate>2023</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=20231124&DB=EPODOC&CC=CN&NR=117115459A$$EHTML$$P50$$Gepo$$Hfree_for_read</linktohtml><link.rule.ids>230,308,780,885,25564,76547</link.rule.ids><linktorsrc>$$Uhttps://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20231124&DB=EPODOC&CC=CN&NR=117115459A$$EView_record_in_European_Patent_Office$$FView_record_in_$$GEuropean_Patent_Office$$Hfree_for_read</linktorsrc></links><search><creatorcontrib>WANG JIN</creatorcontrib><creatorcontrib>ZHOU YANG</creatorcontrib><creatorcontrib>LU GUODONG</creatorcontrib><creatorcontrib>YANG JINGRU</creatorcontrib><creatorcontrib>FANG HEMING</creatorcontrib><title>Sketch recognition method and device based on fusion of three-dimensional sparse convolution and two-dimensional convolution</title><description>The invention discloses a sketch recognition method and device based on fusion of three-dimensional sparse convolution and two-dimensional convolution, and the method comprises the steps: obtaining a sketch image, carrying out the feature extraction of the sketch image, and obtaining the features of the sketch image; meanwhile, sketch point feature extraction and voxelization are conducted, by extracting geometric information of all points on strokes of a sketch, the sketch points are voxelized based on the geometric information, and voxels of the points are obtained; then, carrying out feature extraction on voxels of all points of the sketch to obtain sketch voxel features; and finally, fusing the sketch image features and the sketch voxel features, and carrying out classification identification on the sketch through a classifier. The features of the sketch and all sparse points of the sketch are fully utilized, a large blank area of the sketch is removed through Sketch-SparseVoxelNet, the spatial relation o</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>2023</creationdate><recordtype>patent</recordtype><sourceid>EVB</sourceid><recordid>eNqNjLEKwjAURbs4iPoPzw_oELSIoxTFyUX3EpMbE2zzSpPWxY-3EQfdnC7cczjT7Hm-IypLHRTfvIuOPTWIljVJr0ljcAp0lQGaRmT6kAw2FG0H5No18OmSNYVWdgGk2A9c9-9SSsQH_2hffJ5NjKwDFp-dZcvD_lIec7RcYewpeMSqPAmxEaJYF9vd6h_nBWDHSMY</recordid><startdate>20231124</startdate><enddate>20231124</enddate><creator>WANG JIN</creator><creator>ZHOU YANG</creator><creator>LU GUODONG</creator><creator>YANG JINGRU</creator><creator>FANG HEMING</creator><scope>EVB</scope></search><sort><creationdate>20231124</creationdate><title>Sketch recognition method and device based on fusion of three-dimensional sparse convolution and two-dimensional convolution</title><author>WANG JIN ; ZHOU YANG ; LU GUODONG ; YANG JINGRU ; FANG HEMING</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-epo_espacenet_CN117115459A3</frbrgroupid><rsrctype>patents</rsrctype><prefilter>patents</prefilter><language>chi ; eng</language><creationdate>2023</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>WANG JIN</creatorcontrib><creatorcontrib>ZHOU YANG</creatorcontrib><creatorcontrib>LU GUODONG</creatorcontrib><creatorcontrib>YANG JINGRU</creatorcontrib><creatorcontrib>FANG HEMING</creatorcontrib><collection>esp@cenet</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>WANG JIN</au><au>ZHOU YANG</au><au>LU GUODONG</au><au>YANG JINGRU</au><au>FANG HEMING</au><format>patent</format><genre>patent</genre><ristype>GEN</ristype><title>Sketch recognition method and device based on fusion of three-dimensional sparse convolution and two-dimensional convolution</title><date>2023-11-24</date><risdate>2023</risdate><abstract>The invention discloses a sketch recognition method and device based on fusion of three-dimensional sparse convolution and two-dimensional convolution, and the method comprises the steps: obtaining a sketch image, carrying out the feature extraction of the sketch image, and obtaining the features of the sketch image; meanwhile, sketch point feature extraction and voxelization are conducted, by extracting geometric information of all points on strokes of a sketch, the sketch points are voxelized based on the geometric information, and voxels of the points are obtained; then, carrying out feature extraction on voxels of all points of the sketch to obtain sketch voxel features; and finally, fusing the sketch image features and the sketch voxel features, and carrying out classification identification on the sketch through a classifier. The features of the sketch and all sparse points of the sketch are fully utilized, a large blank area of the sketch is removed through Sketch-SparseVoxelNet, the spatial relation o</abstract><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | |
| ispartof | |
| issn | |
| language | chi ; eng |
| recordid | cdi_epo_espacenet_CN117115459A |
| source | esp@cenet |
| subjects | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING PHYSICS |
| title | Sketch recognition method and device based on fusion of three-dimensional sparse convolution and two-dimensional convolution |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-26T11%3A10%3A57IST&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=WANG%20JIN&rft.date=2023-11-24&rft_id=info:doi/&rft_dat=%3Cepo_EVB%3ECN117115459A%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 |