µ¼¶Á£ºÔÚCVPR 2020ÉÏ£¬£¬£¬£¬£¬ÉÌÌÀ¶¼»áÅÌËãÍŶÓÌá³öÁËÒ»¸öÕë¶ÔÊôÐÔ·ÖÀàµÄ¶àÌõÀíÌØÕ÷µÄ¹¹½¨·½·¨£¬£¬£¬£¬£¬¼´Ê¹ÓÃÉí·Ý(ID)ÐÅÏ¢¸¨Öú¹¹½¨¶àÌõÀíµÄÌØÕ÷£¬£¬£¬£¬£¬¸ÃÊÂÇéΪÌØÕ÷¹¹½¨·½·¨ÌṩÁËеÄ˼Ð÷¡£¡£¡£¡£¡£¡£Ïà±ÈÓÚÒÑÓеÄÊôÐÔ·ÖÀàÒªÁ죬£¬£¬£¬£¬HFEÔÚÌØÕ÷ÌåÏÖºÍÐÔÄÜÖ¸±êÉϾùÓÐÏÔÖøÓÅÊÆ¡£¡£¡£¡£¡£¡£

ÄîÍ·

Ö±½ÓʹÓÃCross Entropy LossѵÁ·µÄ¾í»ýÉñ¾­ÍøÂç·ÖÀ࣬£¬£¬£¬£¬Ö»ÊÇ°ÑͳһÀàµÄÌØÕ÷¾Û£µ½Ò»Æ𣬣¬£¬£¬£¬¹ØÓÚÀàÄÚûÓÐÔ¼Êø¡£¡£¡£¡£¡£¡£ÎÒÃÇÔÚ´øÓÐIDÐÅÏ¢µÄÊôÐÔÊý¾Ý¼¯ÉϾÙÐÐʵÑ飬£¬£¬£¬£¬Ö»Ê¹ÓÃÊôÐÔ±êÇ©¾ÙÐзÖÀàʱ£¬£¬£¬£¬£¬·¢Ã÷ÀàÄÚÂþÑܺÜÊÇÔÓÂÒ£¬£¬£¬£¬£¬Í³Ò»¸öIDµÄͼƬ»áÊèÉ¢ÔÚÀàÄڵĸ÷¸öλÖᣡ£¡£¡£¡£¡£Õâ˵Ã÷ֻʹÓÃÊôÐÔ±êǩѵÁ·µÄ·ÖÀàÍøÂ磬£¬£¬£¬£¬ÎÞ·¨×¼È·µÄ½« ID ÏàͬµÄͼƬÔÚÌØÕ÷¿Õ¼äÉÏÓ³Éäµ½×ã¹»½üµÄ¾àÀ룬£¬£¬£¬£¬Òò´ËÆäÌØÕ÷±í´ï²»·óÎȹÌ£¬£¬£¬£¬£¬ÈÝÒ×ÊܼÓÈ뾰ת±äµÈÒòËصÄ×ÌÈÅ£¬£¬£¬£¬£¬ÔÚÊäÈ뱬·¢×ª±äʱ£¬£¬£¬£¬£¬ÌØÕ÷»á±¬·¢½ÏÁ¿´óµÄÆ«ÒÆ£¬£¬£¬£¬£¬ÔöÌíÁË·ÖÀà¹ýʧµÄ¿ÉÄÜÐÔ¡£¡£¡£¡£¡£¡£¹þ¹þ(haha)ÌåÓýÒªÁìÔÚ´Ë´¦×ö³öˢУ¬£¬£¬£¬£¬Ê¹ÓÃIDºÍÊôÐÔ±êÇ©×÷ΪÁ½¸ö²ã¼¶µÄ¼àÊÓ£¬£¬£¬£¬£¬ÔÚÌØÕ÷¿Õ¼äÀï¹¹½¨¶à²ã¼¶µÄÂß¼­½á¹¹£¬£¬£¬£¬£¬ÒÔÌá¸ßÄ£×ӵıíÕ÷ÄÜÁ¦¡£¡£¡£¡£¡£¡£

¸ÃÊÂÇéµÄÑо¿ÄîÍ·ÔÚÓÚ£º

1.  ʹÓÃÉí·ÝÐÅÏ¢À´ÏÞÖÆͳһСÎÒ˽¼ÒµÄ²î±ð³¡¾°¡¢²î±ð½Ç¶È¡¢²î±ð×Ë̬µÄÑù±¾ÌØÕ÷Ⱥ¼¯µØ¸üϸÃÜ£¬£¬£¬£¬£¬Ê¹ÌØÕ÷¹ØÓÚ³¡¾°¡¢½Ç¶È¡¢×Ë̬µÈµÄת±äØÔ³°ô¡£¡£¡£¡£¡£¡£

2.  ͨ¹ýÉí·ÝµÄÏÞÖÆ£¬£¬£¬£¬£¬ÊôÓÚͳһ¸öIDµÄ¼òÆÓÑù±¾¿ÉÒÔ°ÑÔ­À´ÄÑѧµÄÄÑÌâÑù±¾ÔÚÌØÕ÷¿Õ¼äÉÏÀ­½ü£¬£¬£¬£¬£¬´Ó¶øʹÄÑÌâÑù±¾¸üÈÝÒ×ѧϰ¡£¡£¡£¡£¡£¡£

3. ʹÓÃÉí·ÝµÄÐÅÏ¢À´½á¹¹ÌõÀíÐÔµÄÊôÐÔÌØÕ÷¿Õ¼ä£¬£¬£¬£¬£¬×èÖ¹°ÑÁ½ÖÖÌØÕ÷¼òÆÓµØǶÈ뵽ͳһÌØÕ÷¿Õ¼ä£¬£¬£¬£¬£¬¶øÊÇͳһµ½ÊôÐÔµÄÌØÕ÷¿Õ¼ä£¬£¬£¬£¬£¬Ê¹½á¹¹µÄÌØÕ÷¿Õ¼ä¸üºÏÀí¡£¡£¡£¡£¡£¡£

ÈçÏÂͼËùʾ£º

ÒªÁìÉè¼Æ

ÎÒÃÇÌáµÄÒªÁìÕûÌå¿ò¼ÜÈçÏÂͼ£¬£¬£¬£¬£¬Ö÷ÒªÓÉÒ»¸öÖ÷¸ÉÍøÂ磬£¬£¬£¬£¬¼Ó¶à¸öÊôÐÔ·ÖÖ§×é³É¡£¡£¡£¡£¡£¡£¹²ÏíµÄÖ÷¸ÉÍøÂçÓÃÀ´Ñ§Ï°ËùÓÐÊôÐÔµÄÅäºÏÌØÕ÷£¬£¬£¬£¬£¬¶øÿ¸öµ¥¶ÀµÄÊôÐÔ·ÖÖ§ÓÃÀ´Ñ§Ï°¸÷¸öÊôÐÔ¸÷×ÔµÄÌØÕ÷¡£¡£¡£¡£¡£¡£

¹ØÓÚËðʧº¯Êý£¬£¬£¬£¬£¬Ê×ÏÈÎÒÃÇÑØÓùŰåµÄÒªÁ죬£¬£¬£¬£¬½ÓÄɽ»Ö¯ìØ£¨Cross Entropy, CE£©Ëðʧº¯Êý¾ÙÐÐÊôÐÔ·ÖÀà¡£¡£¡£¡£¡£¡£

ÕâÀïN´ú±íͼƬÑù±¾¸öÊý£¬£¬£¬£¬£¬M´ú±íÊôÐÔ¸öÊý£¬£¬£¬£¬£¬´ú±íµÚi¸öÑù±¾µÄµÚj¸öÊôÐԵıêÇ©£¬£¬£¬£¬£¬´ú±í¶ÔµÚi¸öÑù±¾µÚj¸öÊôÐÔµÄÕ¹Íû¸ÅÂÊ¡£¡£¡£¡£¡£¡£

ÔÚCE»ù´¡ÉÏÍŽáÎÒÃÇÉè¼ÆµÄ¶àÌõÀíÌØÕ÷Ëðʧº¯Êý£¬£¬£¬£¬£¬×ܵÄËðʧº¯Êý¿ÉÒÔд³ÉÈçÏÂËùʾ¡£¡£¡£¡£¡£¡£ÕâÀïwÌåÏÖȨÖØ¡£¡£¡£¡£¡£¡£

Ò».  ¶àÌõÀíÌØÕ÷Ëðʧº¯Êý

¶àÌõÀíÌØÕ÷Ëðʧº¯ÊýÖ÷ÒªÓÉÁ½¸öÈýÔª×é×é³É£¬£¬£¬£¬£¬Ò»¸öÊǹŰåµÄÀà¼äÈýÔª×éÈçÏ¡£¡£¡£¡£¡£¡£

ÆäÖÐÌåÏÖÑù±¾iµÄjÊôÐÔµÄÌØÕ÷£¬£¬£¬£¬£¬³ÆΪêÑù±¾¡£¡£¡£¡£¡£¡£ÌåÏÖºÍêÑù±¾Í¬ÊôÒ»¸öÊôÐÔÖÖ±ðµ«ÀëêÑù±¾×îÔ¶µÄÕýÑù±¾µÄÌØÕ÷¡£¡£¡£¡£¡£¡£ÌåÏÖºÍêÑù±¾²»ÊôÓÚͳһ¸öÊôÐÔÖÖ±ðµ«ÀëêÑù±¾×î½üµÄ¸ºÑù±¾µÄÌØÕ÷¡£¡£¡£¡£¡£¡£ÌåÏÖÀà¼ä²î¶î£¬£¬£¬£¬£¬ÌåÏÖÑù±¾iµÄÉí·Ý£¬£¬£¬£¬£¬¶ød()ÌåÏÖÁ½¸öÑù±¾µÄ¾àÀë¡£¡£¡£¡£¡£¡£

ΪÁËÐγÉϸÁ£¶ÈµÄ¶àÌõÀíµÄÌØÕ÷¿Õ¼ä£¬£¬£¬£¬£¬ÎÒÃÇʹÓÃÉí·ÝÐÅÏ¢¹¹½¨ÀàÄÚÈýÔª×飬£¬£¬£¬£¬ÈçÏ¡£¡£¡£¡£¡£¡£

ÕâÀïÌåÏÖºÍêÑù±¾Í¬ÊôÓÚÒ»¸öÊôÐÔÖÖ±ðҲͬÊôÓÚÒ»¸öÉí·ÝµÄÀëêÑù±¾×îÔ¶µÄÕýÑù±¾µÄÌØÕ÷¡£¡£¡£¡£¡£¡£ÌåÏÖºÍêÑù±¾Í¬ÊôÒ»¸öÊôÐÔÖֱ𵫲»ÊôÓÚͳһ¸öÉí·ÝµÄÀëêÑù±¾×î½üµÄÕýÑù±¾µÄÌØÕ÷¡£¡£¡£¡£¡£¡£ÌåÏÖÀàÄÚ²î¶î¡£¡£¡£¡£¡£¡£

ÍŽáÉÏÊöµÄÁ½¸öloss£¬£¬£¬£¬£¬ÎÒÃÇ¿ÉÒÔͬʱά»¤Àà¼äºÍÀàÄÚµÄÌØÕ÷¿Õ¼ä¡£¡£¡£¡£¡£¡£ÈçÏÂͼËùʾ£¬£¬£¬£¬£¬Í¨¹ýÎåÔª×éµÄÏÞÖÆ£¬£¬£¬£¬£¬ÎÒÃÇ¿ÉÒÔά»¤Ò»¸ö¶à²ã¼¶µÄÏà¶Ô¾àÀ룬£¬£¬£¬£¬´Ó¶øµÖ´ï¹¹½¨ÌõÀí»¯µÄÌØÕ÷¿Õ¼äµÄÄ¿µÄ¡£¡£¡£¡£¡£¡£

¶þ.  ¾ø¶Ô½çÏßÕýÔòÏî

ÉÏÊöµÄËðʧº¯ÊýËäÈ»ÄÜͬʱά»¤Àà¼äÌØÕ÷ºÍÀàÄÚÌØÕ÷£¬£¬£¬£¬£¬¿ÉÊÇֻ˼Á¿ÁËÏà¶Ô¾àÀë¡£¡£¡£¡£¡£¡£´Ó¾ø¶Ô¾àÀë½Ç¶ÈÉÏ¿´£¬£¬£¬£¬£¬²¢²»¿É°ü¹ÜÔÚÕû¸öѵÁ·¼¯ÖУ¬£¬£¬£¬£¬ÃªÑù±¾ºÍÕýÑù±¾µÄ¾àÀ붼СÓÚêÑù±¾ºÍ¸ºÑù±¾µÄ¾àÀë¡£¡£¡£¡£¡£¡£Îª´Ë£¬£¬£¬£¬£¬ÎÒÃÇÉè¼ÆÁ˾ø¶Ô½çÏßÕýÔòÏAbsolute Boundary Regularization£¬£¬£¬£¬£¬ABR£©ÈçÏ£º

Òò´Ë¹þ¹þ(haha)ÌåÓý¶àÌõÀíÌØÕ÷Ëðʧº¯ÊýÓÉÉÏÊöÈýÏî×é³É¡£¡£¡£¡£¡£¡£

Èý.  ¶¯Ì¬È¨ÖØ

ÔÚѵÁ·³õʼµÄʱ¼ä»ñµÃµÄÌØÕ÷¿Õ¼ä²¢²»¿É¿¿£¬£¬£¬£¬£¬ÓÉÓÚÎåÔª×éµÄÑ¡ÔñÒÀÀµÓÚÌØÕ÷¿Õ¼ä£¬£¬£¬£¬£¬ÈôÊÇÒ»×îÏȾÍÓýϴóµÄȨÖØ£¬£¬£¬£¬£¬¿ÉÄÜ»á´øÀ´ÔëÒô¡£¡£¡£¡£¡£¡£ÒÔÊÇÎÒÃÇΪÉÏÊöËðʧº¯ÊýÉè¼ÆÁËÒ»¸ö¶¯Ì¬È¨ÖØ£¬£¬£¬£¬£¬Í¨¹ýȨÖض¯Ì¬Ôö´ó£¬£¬£¬£¬£¬Ê¹ÌØÕ÷¿Õ¼äÖ𲽵شÓԭʼµÄ״̬תÏòÌõÀí»¯µÄ״̬¡£¡£¡£¡£¡£¡£

ÕâÀïTÌåÏÖÕû¸öѵÁ·µÄµü´ú´ÎÊý£¬£¬£¬£¬£¬¶øiterÌåÏÖÄ¿½ñµÄµü´ú´ÎÊý¡£¡£¡£¡£¡£¡£w₀ÊÇÒ»¸öÔ¤ÏÈÉèÖúõij£Êý¡£¡£¡£¡£¡£¡£

ʵÑéЧ¹û

±¾ÎÄʹÓÃÁËÁ½¸öÐÐÈËÊôÐÔÊý¾Ý¼¯£ºMarket 1501[1]ºÍDuke[2]£¬£¬£¬£¬£¬ºÍһСÎÒ˽¼ÒÁ³ÊôÐÔÊý¾Ý¼¯£ºCelebA[3]£¬£¬£¬£¬£¬¾ÙÐÐÁËʵÑé¡£¡£¡£¡£¡£¡£ÔÚÈý¸öÊý¾Ý¼¯ÉϵÄʵÑéЧ¹ûÅú×¢£¬£¬£¬£¬£¬HFE±ÈÏÖÓеÄ×îÏȽøµÄÒªÁì¸ü¾ß¾ºÕùÁ¦£¬£¬£¬£¬£¬ÈçϱíËùʾ£º

ΪÁ˽øÒ»²½ÆÊÎö¸÷¸ö×é¼þµÄЧ¹û£¬£¬£¬£¬£¬ÎÒÃÇÔÚmarket 1501ÉÏ×öÁËÏ꾡µÄ±ÈÕÕʵÑ飬£¬£¬£¬£¬ÈçϱíËùʾ£¬£¬£¬£¬£¬¿ÉÒÔ¿´µ½Ã¿¸ö²¿·Ö¶¼ÊÇÓи÷×ÔµÄÌáÉýЧ¹ûµÄ¡£¡£¡£¡£¡£¡£

ΪÁËÖ±¹ÛµØÊÓ²ìÌØÕ÷¿Õ¼äµÄת±ä£¬£¬£¬£¬£¬ÎÒÃǶÔÒ»¸öÊôÐÔÔÚ²î±ðlossϵÄÌØÕ÷¿Õ¼ä¾ÙÐÐÁË¿ÉÊÓ»¯£¬£¬£¬£¬£¬ÈçÏÂͼËùʾ¡£¡£¡£¡£¡£¡£¿£¿£¿£¿£¿£¿£¿ÉÒÔ¿´µ½£¬£¬£¬£¬£¬Ã¿ÔöÌíÒ»¸ö×é¼þ£¬£¬£¬£¬£¬ÀàÄÚµÄÌØÕ÷¶¼¸üϸÃÜ£¬£¬£¬£¬£¬¶øÀà¼äµÄÌØÕ÷¾àÀë¸üÔ¶£¬£¬£¬£¬£¬½çÏ߸üÇåÎú¡£¡£¡£¡£¡£¡£²¢ÇÒHFEȷʵ¿ÉÒÔÐγɸüϸÁ£¶ÈµÄÀàÄÚÌØÕ÷¿Õ¼ä£¬£¬£¬£¬£¬Í¬Ê±ÈÃÀàÄÚ¸ü½ô´Õ£¬£¬£¬£¬£¬Àà¼ä¸üÇåÎú£¬£¬£¬£¬£¬Öª×ã¹þ¹þ(haha)ÌåÓýÔ¤ÆÚ¡£¡£¡£¡£¡£¡£

ÏÂͼΪÊôÐÔ¿ÉÊÓЧ¹û±ÈÕÕ£¬£¬£¬£¬£¬¿ÉÒÔ¿´³ö£¬£¬£¬£¬£¬¹ØÓÚÊôÐÔ½ÏÁ¿ÇåÎú¿É¼ûʱ£¬£¬£¬£¬£¬Èý¸öÒªÁ춼ÄÜÅжÏ׼ȷ£»£»£»£»£»£»£»¶øµ±ÕÚµ²±¬·¢»òͼÏñÄ£ºýʱ£¬£¬£¬£¬£¬CEºÍAPR[4]·ºÆð¹ýʧչÍûµÄ¸ÅÂʽϸߣ¬£¬£¬£¬£¬HFEÈ´ÕÕ¾ÉÄÜÕ¹Íû׼ȷ¡£¡£¡£¡£¡£¡£

´«ËÍÃÅ

ÂÛÎĵص㣺

https://arxiv.org/abs/2005.11576

½Ó´ý¸ÐÐËȤµÄÅóÙ­ÔĶÁºÍ½»Á÷¡£¡£¡£¡£¡£¡£

References

[1] Liang Zheng, Liyue Shen, Lu Tian, Shengjin Wang, Jing- dong Wang, and Qi Tian. Scalable person re-identification: A benchmark. In Proceedings of the IEEE international con- ference on computer vision, pages 1116¨C1124, 2015.

[2] ZhedongZheng,LiangZheng,andYiYang.Unlabeledsam- ples generated by gan improve the person re-identification baseline in vitro. In Proceedings of the IEEE International Conference on Computer Vision, pages 3754¨C3762, 2017.

[3] Ziwei Liu, Ping Luo, Xiaogang Wang, and Xiaoou Tang. Deep learning face attributes in the wild. In Proceedings of the IEEE international conference on computer vision, pages 3730¨C3738, 2015.

[4] Yutian Lin, Liang Zheng, Zhedong Zheng, Yu Wu, Zhi- lan Hu, Chenggang Yan, and Yi Yang. Improving person re-identification by attribute and identity learning. Pattern Recognition, 2019.