Few-Shot Open-Set Recognition using Meta-Learning

被引:64
作者
Liu, Bo [1 ]
Kang, Hao [2 ]
Li, Haoxiang [2 ]
Hua, Gang [2 ]
Vasconcelos, Nuno [1 ]
机构
[1] UC, San Diego, CA 92093 USA
[2] Wormpex AI Res, Bellevue, WA USA
来源
2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2020) | 2020年
基金
国家重点研发计划;
关键词
D O I
10.1109/CVPR42600.2020.00882
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The problem of open-set recognition is considered. While previous approaches only consider this problem in the context of large-scale classifier training, we seek a unified solution for this and the low-shot classification setting. It is argued that the classic softmax classifier is a poor solution for open-set recognition, since it tends to overfit on the training classes. Randomization is then proposed as a solution to this problem. This suggests the use of meta-learning techniques, commonly used for few-shot classification, for the solution of open-set recognition. A new oPen sEt mEta LEaRning (PEELER) algorithm is then introduced. This combines the random selection of a set of novel classes per episode, a loss that maximizes the posterior entropy for examples of those classes, and a new metric learning formulation based on the Mahalanobis distance. Experimental results show that PEELER achieves state of the art open set recognition performance for both few-shot and large-scale recognition. On CIFAR and miniImageNet, it achieves substantial gains in seen/unseen class detection AUROC for a given seen-class classification accuracy.
引用
收藏
页码:8795 / 8804
页数:10
相关论文
共 37 条
[1]  
[Anonymous], 2017, P IEEE C COMP VIS PA
[2]  
[Anonymous], 2014, arXiv preprint arXiv:1406.2080
[3]   Towards Open Set Deep Networks [J].
Bendale, Abhijit ;
Boult, Terrance E. .
2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, :1563-1572
[4]  
Deng J, 2009, PROC CVPR IEEE, P248, DOI 10.1109/CVPRW.2009.5206848
[5]  
FINN C, 2018, ADV NEURAL INFORM PR
[6]  
Finn C, 2017, PR MACH LEARN RES, V70
[7]  
Ge Zongyuan, 2019, GENERATIVE OPENMAX M
[8]   Dynamic Few-Shot Visual Learning without Forgetting [J].
Gidaris, Spyros ;
Komodakis, Nikos .
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, :4367-4375
[9]   Deep Image Retrieval: Learning Global Representations for Image Search [J].
Gordo, Albert ;
Almazan, Jon ;
Revaud, Jerome ;
Larlus, Diane .
COMPUTER VISION - ECCV 2016, PT VI, 2016, 9910 :241-257
[10]   Deep Residual Learning for Image Recognition [J].
He, Kaiming ;
Zhang, Xiangyu ;
Ren, Shaoqing ;
Sun, Jian .
2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, :770-778