Isosceles Constraints for Person Re-identification

In the existing works of person re-identification (ReID), batch hard triplet loss has achieved great success. However, it only cares about the hardest samples within the batch. For any probe, there are massive mismatched samples (crucial samples) outside the batch which are closer than the matched s...

Ausführliche Beschreibung

Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on image processing : a publication of the IEEE Signal Processing Society. - 1992. - PP(2020) vom: 11. Sept.
1. Verfasser: Xu, Furong (VerfasserIn)
Weitere Verfasser: Ma, Bingpeng, Chang, Hong, Shan, Shiguang
Format: Online-Aufsatz
Sprache:English
Veröffentlicht: 2020
Zugriff auf das übergeordnete Werk:IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
Schlagworte:Journal Article
LEADER 01000caa a22002652 4500
001 NLM314879129
003 DE-627
005 20240229142843.0
007 cr uuu---uuuuu
008 231225s2020 xx |||||o 00| ||eng c
024 7 |a 10.1109/TIP.2020.3020648  |2 doi 
028 5 2 |a pubmed24n1303.xml 
035 |a (DE-627)NLM314879129 
035 |a (NLM)32915737 
040 |a DE-627  |b ger  |c DE-627  |e rakwb 
041 |a eng 
100 1 |a Xu, Furong  |e verfasserin  |4 aut 
245 1 0 |a Isosceles Constraints for Person Re-identification 
264 1 |c 2020 
336 |a Text  |b txt  |2 rdacontent 
337 |a ƒaComputermedien  |b c  |2 rdamedia 
338 |a ƒa Online-Ressource  |b cr  |2 rdacarrier 
500 |a Date Revised 22.02.2024 
500 |a published: Print-Electronic 
500 |a Citation Status Publisher 
520 |a In the existing works of person re-identification (ReID), batch hard triplet loss has achieved great success. However, it only cares about the hardest samples within the batch. For any probe, there are massive mismatched samples (crucial samples) outside the batch which are closer than the matched samples. To reduce the disruptive influence of crucial samples, we propose a novel isosceles contraint for triplet. Theoretically, we show that if a matched pair has equal distance to any one of mismatched sample, the matched pair should be infinitely close. Motivated by this, the isosceles constraint is designed for the two mismatched pairs of each triplet, to restrict some matched pairs with equal distance to different mismatched samples. Meanwhile, to ensure that the distance of mismatched pairs are larger than the matched pairs, margin constraints are necessary. Minimizing the isosceles and margin constraints with respect to the feature extraction network makes the matched pairs closer and the mismatched pairs farther away than the matched ones. By this way, crucial samples are effectively reduced and the performance on ReID is improved greatly. Likewise, our isosceles contraint can be applied to quadruplet as well. Comprehensive experimental evaluations on Market-1501, DukeMTMC-reID and CUHK03 datasets demonstrate the advantages of our isosceles constraint over the related state-of-the-art approaches 
650 4 |a Journal Article 
700 1 |a Ma, Bingpeng  |e verfasserin  |4 aut 
700 1 |a Chang, Hong  |e verfasserin  |4 aut 
700 1 |a Shan, Shiguang  |e verfasserin  |4 aut 
773 0 8 |i Enthalten in  |t IEEE transactions on image processing : a publication of the IEEE Signal Processing Society  |d 1992  |g PP(2020) vom: 11. Sept.  |w (DE-627)NLM09821456X  |x 1941-0042  |7 nnns 
773 1 8 |g volume:PP  |g year:2020  |g day:11  |g month:09 
856 4 0 |u http://dx.doi.org/10.1109/TIP.2020.3020648  |3 Volltext 
912 |a GBV_USEFLAG_A 
912 |a SYSFLAG_A 
912 |a GBV_NLM 
912 |a GBV_ILN_350 
951 |a AR 
952 |d PP  |j 2020  |b 11  |c 09