Deep Adaptive Image Clustering

Image clustering is a crucial but challenging task in machine learning and computer vision. Existing methods often ignore the combination between feature learning and clustering. To tackle this problem, we propose Deep Adaptive Clustering (DAC) that recasts the clustering problem into a binary pairwise-classification framework to judge whether pairs of images belong to the same clusters. In DAC, the similarities are calculated as the cosine distance between label features of images which are generated by a deep convolutional network (ConvNet). By introducing a constraint into DAC, the learned label features tend to be one-hot vectors that can be utilized for clustering images. The main challenge is that the ground-truth similarities are unknown in image clustering. We handle this issue by presenting an alternating iterative Adaptive Learning algorithm where each iteration alternately selects labeled samples and trains the ConvNet. Conclusively, images are automatically clustered based on the label features. Experimental results show that DAC achieves state-of-the-art performance on five popular datasets, e.g., yielding 97.75% clustering accuracy on MNIST, 52.18% on CIFAR-10 and 46.99% on STL-10.

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Image Clustering CIFAR-10 DAC Accuracy 0.522 # 16
NMI 0.4 # 14
Train set Train+Test # 1
ARI 0.301 # 13
Backbone ConvNet # 1

Results from Other Papers

Task Dataset Model Metric Name Metric Value Rank Uses Extra
Training Data
Source Paper Compare
Image Clustering CIFAR-100 DAC Accuracy 0.238 # 13
NMI 0.185 # 11
Train Set Train+Test # 1
Image Clustering ImageNet-10 DAC Accuracy 0.527 # 9
NMI 0.394 # 9
Image Clustering Imagenet-dog-15 DAC Accuracy 0.275 # 9
NMI 0.219 # 9
Image Clustering STL-10 DAC Accuracy 0.470 # 15
NMI 0.366 # 13
Train Split Train+Test # 1
Backbone ConvNet # 1
Image Clustering Tiny-ImageNet DAC Accuracy 0.066 # 7
NMI 0.190 # 7


No methods listed for this paper. Add relevant methods here