Buchumschlag

Deep unsupervised learning of visual similarities

Exemplar learning of visual similarities in an unsupervised manner is a problem of paramount importance to computer vision. In this context, however, the recent breakthrough in deep learning could not yet unfold its full potential. With only a single positive sample, a great imbalance between one po...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Sanakoyeu, Artsiom (VerfasserIn) , Bautista, Miguel (VerfasserIn) , Ommer, Björn (VerfasserIn)
Dokumenttyp: Article (Journal)
Sprache:Englisch
Veröffentlicht: 31 January 2018
In: Pattern recognition
Year: 2018, Jahrgang: 78, Pages: 331-343
DOI:10.1016/j.patcog.2018.01.036
Online-Zugang:Verlag, Volltext: https://doi.org/10.1016/j.patcog.2018.01.036
Verlag, Volltext: http://www.sciencedirect.com/science/article/pii/S0031320318300293
Volltext
Verfasserangaben:Artsiom Sanakoyeu, Miguel A. Bautista, Björn Ommer

MARC

LEADER 00000caa a2200000 c 4500
001 1665801212
003 DE-627
005 20220816153448.0
007 cr uuu---uuuuu
008 190516s2018 xx |||||o 00| ||eng c
024 7 |a 10.1016/j.patcog.2018.01.036  |2 doi 
035 |a (DE-627)1665801212 
035 |a (DE-599)KXP1665801212 
035 |a (OCoLC)1341224818 
040 |a DE-627  |b ger  |c DE-627  |e rda 
041 |a eng 
084 |a 28  |2 sdnb 
100 1 |a Sanakoyeu, Artsiom  |e VerfasserIn  |0 (DE-588)1152280767  |0 (DE-627)1013760638  |0 (DE-576)499782712  |4 aut 
245 1 0 |a Deep unsupervised learning of visual similarities  |c Artsiom Sanakoyeu, Miguel A. Bautista, Björn Ommer 
264 1 |c 31 January 2018 
300 |a 13 
336 |a Text  |b txt  |2 rdacontent 
337 |a Computermedien  |b c  |2 rdamedia 
338 |a Online-Ressource  |b cr  |2 rdacarrier 
500 |a Gesehen am 16.05.2019 
520 |a Exemplar learning of visual similarities in an unsupervised manner is a problem of paramount importance to computer vision. In this context, however, the recent breakthrough in deep learning could not yet unfold its full potential. With only a single positive sample, a great imbalance between one positive and many negatives, and unreliable relationships between most samples, training of Convolutional Neural networks is impaired. In this paper we use weak estimates of local similarities and propose a single optimization problem to extract batches of samples with mutually consistent relations. Conflicting relations are distributed over different batches and similar samples are grouped into compact groups. Learning visual similarities is then framed as a sequence of categorization tasks. The CNN then consolidates transitivity relations within and between groups and learns a single representation for all samples without the need for labels. The proposed unsupervised approach has shown competitive performance on detailed posture analysis and object classification. 
650 4 |a Deep learning 
650 4 |a Human pose analysis 
650 4 |a Object retrieval 
650 4 |a Self-supervised learning 
650 4 |a Visual similarity learning 
700 1 |a Bautista, Miguel  |e VerfasserIn  |0 (DE-588)115223787X  |0 (DE-627)1013721810  |0 (DE-576)49975705X  |4 aut 
700 1 |a Ommer, Björn  |d 1981-  |e VerfasserIn  |0 (DE-588)1034893106  |0 (DE-627)746457510  |0 (DE-576)382507916  |4 aut 
773 0 8 |i Enthalten in  |t Pattern recognition  |d Amsterdam : Elsevier, 1968  |g 78(2018), Seite 331-343  |h Online-Ressource  |w (DE-627)265784131  |w (DE-600)1466343-0  |w (DE-576)101177364  |7 nnas  |a Deep unsupervised learning of visual similarities 
773 1 8 |g volume:78  |g year:2018  |g pages:331-343  |g extent:13  |a Deep unsupervised learning of visual similarities 
856 4 0 |u https://doi.org/10.1016/j.patcog.2018.01.036  |x Verlag  |x Resolving-System  |3 Volltext 
856 4 0 |u http://www.sciencedirect.com/science/article/pii/S0031320318300293  |x Verlag  |3 Volltext 
951 |a AR 
992 |a 20190516 
993 |a Article 
994 |a 2018 
998 |g 1034893106  |a Ommer, Björn  |m 1034893106:Ommer, Björn  |d 700000  |d 708000  |e 700000PO1034893106  |e 708000PO1034893106  |k 0/700000/  |k 1/700000/708000/  |p 3  |y j 
998 |g 115223787X  |a Bautista, Miguel  |m 115223787X:Bautista, Miguel  |d 700000  |d 708000  |e 700000PB115223787X  |e 708000PB115223787X  |k 0/700000/  |k 1/700000/708000/  |p 2 
998 |g 1152280767  |a Sanakoyeu, Artsiom  |m 1152280767:Sanakoyeu, Artsiom  |d 110000  |e 110000PS1152280767  |k 0/110000/  |p 1  |x j 
999 |a KXP-PPN1665801212  |e 3476472744 
BIB |a Y 
SER |a journal 
JSO |a {"physDesc":[{"extent":"13 S."}],"relHost":[{"physDesc":[{"extent":"Online-Ressource"}],"id":{"zdb":["1466343-0"],"eki":["265784131"]},"origin":[{"dateIssuedDisp":"1968-","publisher":"Elsevier","dateIssuedKey":"1968","publisherPlace":"Amsterdam"}],"part":{"year":"2018","pages":"331-343","volume":"78","text":"78(2018), Seite 331-343","extent":"13"},"pubHistory":["1.1968/69 - 48.2015; Vol. 49.2016 -"],"recId":"265784131","language":["eng"],"corporate":[{"role":"isb","display":"Pattern Recognition Society","roleDisplay":"Herausgebendes Organ"}],"type":{"media":"Online-Ressource","bibl":"periodical"},"note":["Gesehen am 31.05.23"],"disp":"Deep unsupervised learning of visual similaritiesPattern recognition","title":[{"subtitle":"the journal of the Pattern Recognition Society","title":"Pattern recognition","title_sort":"Pattern recognition"}]}],"origin":[{"dateIssuedDisp":"31 January 2018","dateIssuedKey":"2018"}],"id":{"eki":["1665801212"],"doi":["10.1016/j.patcog.2018.01.036"]},"name":{"displayForm":["Artsiom Sanakoyeu, Miguel A. Bautista, Björn Ommer"]},"type":{"bibl":"article-journal","media":"Online-Ressource"},"note":["Gesehen am 16.05.2019"],"recId":"1665801212","language":["eng"],"title":[{"title":"Deep unsupervised learning of visual similarities","title_sort":"Deep unsupervised learning of visual similarities"}],"person":[{"role":"aut","display":"Sanakoyeu, Artsiom","roleDisplay":"VerfasserIn","given":"Artsiom","family":"Sanakoyeu"},{"roleDisplay":"VerfasserIn","display":"Bautista, Miguel","role":"aut","family":"Bautista","given":"Miguel"},{"roleDisplay":"VerfasserIn","display":"Ommer, Björn","role":"aut","family":"Ommer","given":"Björn"}]} 
SRT |a SANAKOYEUADEEPUNSUPE3120 

Ähnliche Einträge