{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,26]],"date-time":"2026-03-26T19:02:54Z","timestamp":1774551774435,"version":"3.50.1"},"reference-count":40,"publisher":"Wiley","issue":"7","license":[{"start":{"date-parts":[[2017,10,13]],"date-time":"2017-10-13T00:00:00Z","timestamp":1507852800000},"content-version":"vor","delay-in-days":12,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Computer Graphics Forum"],"published-print":{"date-parts":[[2017,10]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>Many image editing applications rely on the analysis of image patches. In this paper, we present a method to analyze patches by embedding them to a vector space, in which the Euclidean distance reflects patch similarity. Inspired by Word2Vec, we term our approach<jats:italic>Patch2Vec<\/jats:italic>. However, there is a significant difference between words and patches. Words have a fairly small and well defined dictionary. Image patches, on the other hand, have no such dictionary and the number of different patch types is not well defined. The problem is aggravated by the fact that each patch might contain several objects and textures. Moreover, Patch2Vec should be universal because it must be able to map never\u2010seen\u2010before texture to the vector space. The mapping is learned by analyzing the distribution of all natural patches. We use Convolutional Neural Networks (CNN) to learn Patch2Vec. In particular, we train a CNN on labeled images with a triplet\u2010loss objective function. The trained network encodes a given patch to a 128D vector. Patch2Vec is evaluated visually, qualitatively, and quantitatively. We then use several variants of an interactive single\u2010click image segmentation algorithm to demonstrate the power of our method.<\/jats:p>","DOI":"10.1111\/cgf.13284","type":"journal-article","created":{"date-parts":[[2017,10,13]],"date-time":"2017-10-13T09:25:12Z","timestamp":1507886712000},"page":"183-194","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":13,"title":["Patch2Vec: Globally Consistent Image Patch Representation"],"prefix":"10.1111","volume":"36","author":[{"given":"O.","family":"Fried","sequence":"first","affiliation":[{"name":"Princeton University"}]},{"given":"S.","family":"Avidan","sequence":"additional","affiliation":[{"name":"Tel\u2010Aviv University"}]},{"given":"D.","family":"Cohen\u2010Or","sequence":"additional","affiliation":[{"name":"Tel\u2010Aviv University"}]}],"member":"311","published-online":{"date-parts":[[2017,10,13]]},"reference":[{"key":"e_1_2_7_2_2","unstructured":"AmosB. LudwiczukB. SatyanarayananM.:Open\u2010Face: A general\u2010purpose face recognition library with mobile applications. Tech. rep. CMU\u2010CS\u201016\u2010118 2016. 4 12"},{"key":"e_1_2_7_3_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2010.161"},{"key":"e_1_2_7_4_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2012.120"},{"key":"e_1_2_7_5_2","doi-asserted-by":"crossref","unstructured":"BagonS. BoimanO. IraniM.:What is a good image segment? a unified approach to segment extraction. In ECCV(2008) pp.30\u201344. 3","DOI":"10.1007\/978-3-540-88693-8_3"},{"key":"e_1_2_7_6_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2005.38"},{"key":"e_1_2_7_7_2","doi-asserted-by":"publisher","DOI":"10.1145\/2508363.2508421"},{"key":"e_1_2_7_8_2","first-page":"549","volume-title":"What's the Point: Semantic Segmentation with Point Supervision","author":"Bearman A.","year":"2016"},{"key":"e_1_2_7_9_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-008-0191-z"},{"key":"e_1_2_7_10_2","doi-asserted-by":"publisher","DOI":"10.1145\/1531326.1531330"},{"key":"e_1_2_7_11_2","unstructured":"BonetJ. S. D. ViolaP.A.:Texture recognition using a non\u2010parametric multi\u2010scale statistical model. In CVPR(1998) pp.641\u2013647. 3"},{"key":"e_1_2_7_12_2","doi-asserted-by":"publisher","DOI":"10.1007\/s41095-016-0064-2"},{"key":"e_1_2_7_13_2","doi-asserted-by":"crossref","unstructured":"BarnesC. ZhangF.\u2010L. LouL. WuX. HuS.\u2010M.:Patchtable: Efficient patch queries for large datasets and applications. In ACM Trans. Graph. (Proc. SIGGRAPH)(Aug.2015). 2","DOI":"10.1145\/2766934"},{"key":"e_1_2_7_14_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.461"},{"key":"e_1_2_7_15_2","doi-asserted-by":"crossref","unstructured":"CimpoiM. MajiS. VedaldiA.:Deep filter banks for texture recognition and segmentation. In CVPR(June2015) pp.3828\u20133836. 3","DOI":"10.1109\/CVPR.2015.7299007"},{"key":"e_1_2_7_16_2","doi-asserted-by":"crossref","unstructured":"DoerschC. GuptaA. EfrosA.A.:Unsupervised visual representation learning by context prediction. In ICCV(2015). 4 5 6","DOI":"10.1109\/ICCV.2015.167"},{"key":"e_1_2_7_17_2","doi-asserted-by":"crossref","unstructured":"Doll\u00e1rP. ZitnickC.L.:Structured forests for fast edge detection. In ICCV(2013). 7 11","DOI":"10.1109\/ICCV.2013.231"},{"key":"e_1_2_7_18_2","doi-asserted-by":"crossref","unstructured":"EfrosA.A. LeungT.K.:Texture synthesis by non\u2010parametric sampling. In ICCV(1999) pp.1033\u20131038. 3","DOI":"10.1109\/ICCV.1999.790383"},{"key":"e_1_2_7_19_2","doi-asserted-by":"publisher","DOI":"10.1145\/1882261.1866171"},{"key":"e_1_2_7_20_2","first-page":"229","volume-title":"Proceedings of the 22Nd Annual Conference on Computer Graphics and Interactive Techniques","author":"Heeger D.J.","year":"1995"},{"key":"e_1_2_7_21_2","doi-asserted-by":"publisher","DOI":"10.1145\/2508363.2508381"},{"key":"e_1_2_7_22_2","doi-asserted-by":"publisher","DOI":"10.1016\/0031-3203(91)90143-S"},{"key":"e_1_2_7_23_2","doi-asserted-by":"publisher","DOI":"10.1038\/290091a0"},{"key":"e_1_2_7_24_2","doi-asserted-by":"publisher","DOI":"10.1007\/BF00133570"},{"key":"e_1_2_7_25_2","doi-asserted-by":"crossref","unstructured":"LinT.\u2010Y. MajiS.:Visualizing and understanding deep texture representations. In CVPR(June2016). 3","DOI":"10.1109\/CVPR.2016.305"},{"key":"e_1_2_7_26_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2008.168"},{"key":"e_1_2_7_27_2","unstructured":"LockermanY.D. SauvageB. All\u00e8greR. DischlerJ. DorseyJ. RushmeierH.:Multi\u2010scale label\u2010map extraction for texture synthesis. ACM Trans. Graph. (07\/20162016). 3"},{"key":"e_1_2_7_28_2","unstructured":"LongJ. ShelhamerE. DarrellT.:Fully convolutional networks for semantic segmentation. In CVPR(June2015). 3 4 5 6 7"},{"key":"e_1_2_7_29_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2004.1273918"},{"key":"e_1_2_7_30_2","unstructured":"MikolovT. YihW. ZweigG.:Linguistic regularities in continuous space word representations. In Human Language Technologies: Conference of the North American Chapter of the Association of Computational Linguistics(2013) pp.746\u2013751. 2"},{"key":"e_1_2_7_31_2","doi-asserted-by":"crossref","unstructured":"Nathan Silberman Derek HoiemP.K. FergusR.:Indoor segmentation and support inference from rgbd images. In ECCV(2012). 9 11","DOI":"10.1007\/978-3-642-33715-4_54"},{"key":"e_1_2_7_32_2","doi-asserted-by":"publisher","DOI":"10.1109\/TSMC.1979.4310076"},{"key":"e_1_2_7_33_2","doi-asserted-by":"publisher","DOI":"10.1109\/34.761261"},{"key":"e_1_2_7_34_2","doi-asserted-by":"publisher","DOI":"10.1145\/1015706.1015720"},{"key":"e_1_2_7_35_2","doi-asserted-by":"crossref","unstructured":"SchroffF. KalenichenkoD. PhilbinJ.:Facenet: A unified embedding for face recognition and clustering. In CVPR(June2015) pp.815\u2013823. 3 4 12","DOI":"10.1109\/CVPR.2015.7298682"},{"key":"e_1_2_7_36_2","doi-asserted-by":"crossref","unstructured":"SzegedyC. LiuW. JiaY. SermanetP. ReedS. AnguelovD. ErhanD. VanhouckeV. RabinovichA.:Going deeper with convolutions. In CVPR(June2015) pp.1\u20139. 12","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"e_1_2_7_37_2","doi-asserted-by":"crossref","unstructured":"Simo\u2010SerraE. TrullsE. FerrazL. KokkinosJ. FuaP. Moreno\u2010NoguerF.:Discriminative learning of deep convolutional feature point descriptors. In ICCV(2015) pp.118\u2013126. 2","DOI":"10.1109\/ICCV.2015.22"},{"key":"e_1_2_7_38_2","doi-asserted-by":"publisher","DOI":"10.1145\/2980179.2980253"},{"issue":"1","key":"e_1_2_7_39_2","first-page":"2287","article-title":"Stereo matching by training a convolutional neural network to compare image patches","volume":"17","author":"\u017dbontar J.","year":"2016","journal-title":"J. Mach. Learn. Res"},{"key":"e_1_2_7_40_2","doi-asserted-by":"crossref","unstructured":"VarmaM. ZissermanA.:Texture classification: Are filter banks necessary? In CVPR(2003) pp.691\u2013698. 3","DOI":"10.1109\/CVPR.2003.1211534"},{"key":"e_1_2_7_41_2","doi-asserted-by":"crossref","unstructured":"XuN. PriceB. CohenS. YangJ. HuangT.:Deep interactive object selection. In CVPR(June2016) pp.373\u2013381. 3","DOI":"10.1109\/CVPR.2016.47"}],"container-title":["Computer Graphics Forum"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.wiley.com\/onlinelibrary\/tdm\/v1\/articles\/10.1111%2Fcgf.13284","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/cgf.13284","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,26]],"date-time":"2025-06-26T13:21:25Z","timestamp":1750944085000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/cgf.13284"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2017,10]]},"references-count":40,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2017,10]]}},"alternative-id":["10.1111\/cgf.13284"],"URL":"https:\/\/doi.org\/10.1111\/cgf.13284","archive":["Portico"],"relation":{},"ISSN":["0167-7055","1467-8659"],"issn-type":[{"value":"0167-7055","type":"print"},{"value":"1467-8659","type":"electronic"}],"subject":[],"published":{"date-parts":[[2017,10]]},"assertion":[{"value":"2017-10-13","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}