Discriminatively Boosted Image Clustering with Fully Convolutional Auto-Encoders

23 Mar 2017 Fengfu Li Hong Qiao Bo Zhang Xuanyang Xi

Traditional image clustering methods take a two-step approach, feature learning and clustering, sequentially. However, recent research results demonstrated that combining the separated phases in a unified framework and training them jointly can achieve a better performance... (read more)

PDF Abstract

Datasets


Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Image Clustering coil-100 DBC NMI 0.905 # 7
Accuracy 0.775 # 5
Image Clustering Coil-20 DBC NMI 0.895 # 3
Accuracy 0.793 # 3
Image Clustering MNIST-full DBC NMI 0.917 # 7
Accuracy 0.964 # 7
Image Clustering USPS DBC NMI 0.724 # 12
Accuracy 0.743 # 10

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet