Name
Abstract | ||
|---|---|---|
The clustering methods have recently absorbed even-increasing attention in learning and vision. Deep clustering combines embedding and clustering together to obtain optimal embedding subspace for clustering, which can be more effective compared with conventional clustering methods. In this paper, we propose a joint learning framework for discriminative embedding and spectral clustering. We first devise a dual autoencoder network, which enforces the reconstruction constraint for the latent representations and their noisy versions, to embed the inputs into a latent space for clustering. As such the learned latent representations can be more robust to noise. Then the mutual information estimation is utilized to provide more discriminative information from the inputs. Furthermore, a deep spectral clustering method is applied to embed the latent representations into the eigenspace and subsequently clusters them, which can fully exploit the relationship between inputs to achieve optimal clustering results. Experimental results on bench-mark datasets show that our method can significantly outperform state-of-the-art clustering approaches. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1109/CVPR.2019.00419 | 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2019) |
Field | DocType | Volume |
Spectral clustering,Embedding,Autoencoder,Subspace topology,Mutual information,Artificial intelligence,Cluster analysis,Discriminative model,Eigenvalues and eigenvectors,Machine learning,Mathematics | Journal | abs/1904.13113 |
ISSN | Citations | PageRank |
1063-6919 | 17 | 0.62 |
References | Authors | |
0 | 5 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Xu Yang | 1 | 45 | 8.16 |
| Cheng Deng | 2 | 1283 | 85.48 |
| Feng Zheng | 3 | 369 | 31.93 |
| Junchi Yan | 4 | 891 | 83.36 |
| Wei Liu | 5 | 4041 | 204.19 |