000223771 001__ 223771
000223771 005__ 20190317000609.0
000223771 020__ $$a978-3-319-46604-0
000223771 0247_ $$2doi$$a10.1007/978-3-319-46604-0_52
000223771 037__ $$aCONF
000223771 245__ $$aVisual Link Retrieval in a Database of Paintings
000223771 269__ $$a2016
000223771 260__ $$c2016
000223771 336__ $$aConference Papers
000223771 520__ $$aThis paper examines how far state-of-the-art machine vision algorithms can be used to retrieve common visual patterns shared by series of paintings. The research of such visual patterns, central to Art History Research, is challenging because of the diversity of similarity criteria that could relevantly demonstrate genealogical links. We design a methodology and a tool to annotate efficiently clusters of similar paintings and test various algorithms in a retrieval task. We show that pretrained convolutional neural network can perform better for this task than other machine vision methods aimed at photograph analysis. We also show that retrieval performance can be significantly improved by fine-tuning a network specifically for this task.
000223771 6531_ $$aComputer Vision
000223771 6531_ $$aMachine Learning
000223771 700__ $$0247542$$g211479$$aSeguin, Benoît Laurent Auguste
000223771 700__ $$aStriolo, Carlota
000223771 700__ $$0248191$$g246947$$adi Lenardo, Isabella
000223771 700__ $$aKaplan, Frédéric$$g174121$$0240415
000223771 7112_ $$dSeptember, 2016$$cAmsterdam$$aVISART Workshop, ECCV
000223771 8564_ $$uhttps://infoscience.epfl.ch/record/223771/files/Seguin%20et%20al.%20-%202016%20-%20Visual%20Link%20Retrieval%20in%20a%20Database%20of%20Paintings.pdf$$zPreprint$$s4618691$$yPreprint
000223771 909C0 $$xU12632$$0252465$$pDHLAB
000223771 909CO $$ooai:infoscience.tind.io:223771$$qGLOBAL_SET$$pconf$$pCDH
000223771 917Z8 $$x211479
000223771 917Z8 $$x211479
000223771 937__ $$aEPFL-CONF-223771
000223771 973__ $$rREVIEWED$$sPUBLISHED$$aOTHER
000223771 980__ $$aCONF