Name
Abstract | ||
|---|---|---|
For large-scale visual search, highly compressed yet meaningful representations of images are essential. Structured vector quantizers based on product quantization and its variants are usually employed to achieve such compression while minimizing the loss of accuracy. Yet, unlike binary hashing schemes, these unsupervised methods have not yet benefited from the supervision, end-to-end learning and novel architectures ushered in by the deep learning revolution. We hence propose herein a novel method to make deep convolutional neural networks produce supervised, compact, structured binary codes for visual search. Our method makes use of a novel block-softmax nonlinearity and of batch-based entropy losses that together induce structure in the learned encodings. We show that our method outperforms state-of-the-art compact representations based on deep hashing or structured quantization in single and cross-domain category retrieval, instance retrieval and classification. We make our code and models publicly available online. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1109/ICCV.2017.96 | 2017 IEEE International Conference on Computer Vision (ICCV) |
Keywords | DocType | Volume |
supervised codes,compact codes,structured binary codes,SuBiC,image search,large-scale visual search,structured vector quantizers,product quantization,binary hashing schemes,end-to-end learning,deep learning revolution,deep convolutional neural networks,block-softmax nonlinearity losses,batch-based entropy losses,encodings learning,image representations | Conference | abs/1708.02932 |
Issue | ISSN | ISBN |
1 | 1550-5499 | 978-1-5386-1033-6 |
Citations | PageRank | References |
17 | 0.78 | 31 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Himalaya Jain | 1 | 17 | 0.78 |
| Joaquin Zepeda | 2 | 19 | 1.81 |
| Patrick Pérez | 3 | 6529 | 391.34 |
| Rémi Gribonval | 4 | 55 | 4.74 |