Abstract
Since the number of 3D models is rapidly increasing, extracting better feature descriptors to represent 3D models is very challenging for effective 3D model retrieval. There are some problems in existing 3D model representation approaches. For example, many of them focus on the direct extraction of features or transforming 3D models into 2D images for feature extraction, which cannot effectively represent 3D models. In this paper, we propose a novel 3D model feature representation method that is a kind of voxelization method. It is based on the space-based concept, namely CSS (Cube of Space Sampling). The CSS method uses cube space 3D model sampling to extract global and local features of 3D models. The experiments using the ESB dataset show that the proposed method to extract the voxel-based features can provide better classification accuracy than SVM and comparable retrieval results using the state-of-the-art 3D model feature representation method.
Highlights
In the last few years, 3D images and models have gradually extended to different applications and become more popular, such as in Computer-Aided Design (CAD) [1], molecular biology, virtual reality, video and computer games, movies etc
We propose a novel approach for 3D model re-sampling, namely CSS (Cube of Space Sampling)
It is a kind of voxelization method and can be transformed into the voxel-based feature for 3D model classification and retrieval
Summary
In the last few years, 3D images and models have gradually extended to different applications and become more popular, such as in Computer-Aided Design (CAD) [1], molecular biology, virtual reality, video and computer games, movies etc. These applications usually require managing a large number of 3D models. Some of the existing feature extraction methods are only based on 2D images, which use many different views per image to represent a 3D model. Compared with some of these methods, image-based approaches have shown to produce some promising results, but they are not effective over various types of 3D models. It is worth mentioning that we believe that t2h.e1.v3oDxeMl-obdaesleRdetfreiaetvuarle approach has great potential and robustness for 3D model ret3imt4pdcsdanrrDioianariifomevsoocrnddvmcrticrdiidereaITlmseoPWoiancllhes5mhn,rtdaarsda,leCaheseteyimpranitnitisrSleoplernseid3aeosSaind.ptsedpDtd,uot,Itteirvheehnocebtonoatmcfouelanuhaoftasltmtbi)ica.etordtfav;eohhmbydaa(ndfaeu;i2oioltsedsliuansi)arpyvtllcotiptglnAf.eiaeoarooeoadepxclntwmfuanlphtepory,tret.eoieauueoit3nx(r(evhtrrrmDs4dpgiviiieeoum)isdeeaasincmreCgresolsteehix;sirurarmniopoutergenleartdorneceseasasfseubeeo:snncoscacl:aulrtutiefoaoremzartssnestnrmscdaeltiiatihanfovdmnramipnfteenergteifapsamisosrioaolslsrusymlncaatpiin;uflunsttritooeo;ahiomirsgsotelrSnneinelyesndapeosids,lsn;c:awbosonaotrdcoreureclioosngaitfopatsg:tafittwlunlehhtacSnnochmdfutieeei3tmneisee,ecloaettlcanittneyiofntnnoihtcotunoittlna3epeoaly.rrtnltcDx,rnoenomelrot2esddudwre.svmtaaoustcpsiiimrciv.cdnnocoei)teeeagdeeen(aswvrltd3edivtihiatn)hlsh,nhivsorsegReeog,eeaudw;tlosngce3eppsmobsl.t,sDrroatu:rentooo(gbhe1scsmvpeaeltch)oawainoloodtroCtscedrmdeefuehfseoiddee3dssscaprla:eDttrwsnlirftnreten.euighscoembtaeretiSsrcutesnsoskeuacts;ecdcin(eronosatfgeerifrsonnfiotolo:e3asinonsbmoxDitiddensoef-s; 2Tt.hhLeeiritdeefreoaarteuo,frien“bRoueridvldeierowtnocfeu,llryeudseemoofntesntr”a(tei.eth.,ims aoddvifaynttahgeemotfo3Dbemcoomdeelas, new model or i.e., to achieve reuse). re-use, 2i.t1i.s3vDerMy oimdepl oRrettarnietvtaol accurately classify 3D models for the system in order to find the mPaost s3iDmmilaordmelordeetlriteovaalqrueesreya.rch focuses on the following issues: correctness (good discrimination); automation; robustness; and calculation speed, etc. (1) Correctness: to correctly identifying the degree of similarity between 3D models, so that users can find similar models. (2) Automation: using more sophisticated methods to complete certain processes automatically (such as comparison, alignment, etc.) (3) Robustness: this can be
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
