ForestHash: Semantic hashing with shallow random forests and tiny convolutional networks
Qiu Q, Lezama J, Bronstein AM, Sapiro G. 2018. ForestHash: Semantic hashing with shallow random forests and tiny convolutional networks. European Conference on Computer Vision. ECCV: European Conference on Computer Vision, LNCS, vol. 11206.
Download
No fulltext has been uploaded. References only!
Conference Paper
| Published
| English
Scopus indexed
Author
Qiu, Qiang;
Lezama, José;
Bronstein, Alex M.ISTA
;
Sapiro, Guillermo

Series Title
LNCS
Abstract
In this paper, we introduce a random forest semantic hashing scheme that embeds tiny convolutional neural networks (CNN) into shallow random forests. A binary hash code for a data point is obtained by a set of decision trees, setting ‘1’ for the visited tree leaf, and ‘0’ for the rest. We propose to first randomly group arriving classes at each tree split node into two groups, obtaining a significantly simplified two-class classification problem that can be a handled with a light-weight CNN weak learner. Code uniqueness is achieved via the random class grouping, whilst code consistency is achieved using a low-rank loss in the CNN weak learners that encourages intra-class compactness for the two random class groups. Finally, we introduce an information-theoretic approach for aggregating codes of individual trees into a single hash code, producing a near-optimal unique hash for each class. The proposed approach significantly outperforms state-of-the-art hashing methods for image retrieval tasks on large-scale public datasets, and is comparable to image classification methods while utilizing a more compact, efficient and scalable representation. This work proposes a principled and robust procedure to train and deploy in parallel an ensemble of light-weight CNNs, instead of simply going deeper.
Publishing Year
Date Published
2018-10-09
Proceedings Title
European Conference on Computer Vision
Publisher
Springer Nature
Volume
11206
Issue
Part II
Conference
ECCV: European Conference on Computer Vision
Conference Location
Munich, Germany
Conference Date
2018-09-08 – 2018-09-14
ISBN
ISSN
eISSN
IST-REx-ID
Cite this
Qiu Q, Lezama J, Bronstein AM, Sapiro G. ForestHash: Semantic hashing with shallow random forests and tiny convolutional networks. In: European Conference on Computer Vision. Vol 11206. Springer Nature; 2018. doi:10.1007/978-3-030-01216-8_27
Qiu, Q., Lezama, J., Bronstein, A. M., & Sapiro, G. (2018). ForestHash: Semantic hashing with shallow random forests and tiny convolutional networks. In European Conference on Computer Vision (Vol. 11206). Munich, Germany: Springer Nature. https://doi.org/10.1007/978-3-030-01216-8_27
Qiu, Qiang, José Lezama, Alex M. Bronstein, and Guillermo Sapiro. “ForestHash: Semantic Hashing with Shallow Random Forests and Tiny Convolutional Networks.” In European Conference on Computer Vision, Vol. 11206. Springer Nature, 2018. https://doi.org/10.1007/978-3-030-01216-8_27.
Q. Qiu, J. Lezama, A. M. Bronstein, and G. Sapiro, “ForestHash: Semantic hashing with shallow random forests and tiny convolutional networks,” in European Conference on Computer Vision, Munich, Germany, 2018, vol. 11206, no. Part II.
Qiu Q, Lezama J, Bronstein AM, Sapiro G. 2018. ForestHash: Semantic hashing with shallow random forests and tiny convolutional networks. European Conference on Computer Vision. ECCV: European Conference on Computer Vision, LNCS, vol. 11206.
Qiu, Qiang, et al. “ForestHash: Semantic Hashing with Shallow Random Forests and Tiny Convolutional Networks.” European Conference on Computer Vision, vol. 11206, no. Part II, Springer Nature, 2018, doi:10.1007/978-3-030-01216-8_27.