|Table of Contents|

[1] Zuo Xin, Luo Limin, Shen Jifeng, et al. Learning compact binary codebased on multiple heterogeneous features [J]. Journal of Southeast University (English Edition), 2013, 29 (4): 372-378. [doi:10.3969/j.issn.1003-7985.2013.04.004]
Copy

Learning compact binary codebased on multiple heterogeneous features()
基于多源异质特征的紧致二进制编码学习
Share:

Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]

Volumn:
29
Issue:
2013 4
Page:
372-378
Research Field:
Computer Science and Engineering
Publishing date:
2013-12-20

Info

Title:
Learning compact binary codebased on multiple heterogeneous features
基于多源异质特征的紧致二进制编码学习
Author(s):
Zuo Xin1 2 Luo Limin1 Shen Jifeng3 Yu Hualong2
1Laboratory of Image Science and Technology, Southeast University, Nanjing 210096, China
2 School of Computer Science and Engineering, Jiangsu University of Science and Technology, Zhenjiang 212003, China
3School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, China
左欣1 2 罗立民1 沈继锋3 于化龙2
1东南大学影像科学与技术实验室, 南京 210096; 2江苏科技大学计算机科学与工程学院, 镇江212003; 3江苏大学电气信息工程学院, 镇江212013
Keywords:
hashing code linear discriminate analysis asymmetric boosting heterogeneous feature
哈希编码 线性判别分析 非对称boosting 异质特征
PACS:
TP391.3
DOI:
10.3969/j.issn.1003-7985.2013.04.004
Abstract:
A novel hashing method based on multiple heterogeneous features is proposed to improve the accuracy of the image retrieval system. First, it leverages the imbalanced distribution of the similar and dissimilar samples in the feature space to boost the performance of each weak classifier in the asymmetric boosting framework. Then, the weak classifier based on a novel linear discriminate analysis(LDA)algorithm which is learned from the subspace of heterogeneous features is integrated into the framework. Finally, the proposed method deals with each bit of the code sequentially, which utilizes the samples misclassified in each round in order to learn compact and balanced code. The heterogeneous information from different modalities can be effectively complementary to each other, which leads to much higher performance. The experimental results based on the two public benchmarks demonstrate that this method is superior to many of the state-of-the-art methods. In conclusion, the performance of the retrieval system can be improved with the help of multiple heterogeneous features and the compact hash codes which can be learned by the imbalanced learning method.
为了提高图像检索系统的精度, 提出了一种基于多种异质特征的新颖哈希函数学习方法.该方法首先利用特征空间中相似样本与非相似样本分布的不平衡性来提升每个弱分类器的性能, 从而建立非对称的Boosting框架;然后将一种基于异质特征子空间学习的线性判别弱分类器融入该框架下, 并利用每轮算法中的误判样本的信息来依次学习紧致且平衡的哈希编码.该方法能有效地融合具有互补功能的不同模态的信息, 实现了检索系统的性能提升.在2个公开数据集上的实验结果表明该方法优于其他算法, 由此看出增加多源异质特征和利用不平衡性学习紧致哈希编码都可以大大提高图像检索的精度.

References:

[1] Silpa-Anan C, Hartley R. Optimised KD-trees for fast image descriptor matching[C]//IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, AK, USA, 2008: 1-8.
[2] Uhlmann J K. Satisfying general proximity/similarity queries with metric trees[J]. Information Processing Letters, 1991, 40(4):175-179.
[3] Gionis A, Indyk P, Motwani R. Similarity search in high dimensions via hashing[C]//Proceedings of the 25th International Conference on Very Large Data Bases. Edinburgh, Scotland, 1999: 518-529.
[4] Lv Q, Josephson W, Wang Z, et al. Multi-probe LSH: efficient indexing for high-dimensional similarity search[C]//Proceedings of the 33rd International Conference on Very Large Data Bases. Vienna, Austria, 2007: 950-961.
[5] Kulis B, Grauman K. Kernelized locality-sensitive hashing[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(6): 1092-1104.
[6] Gong Y, Lazebnik S. Iterative quantization: a procrustean approach to learning binary codes[C]//IEEE Conference on Computer Vision and Pattern Recognition. Providence, RI, USA, 2011: 817-824.
[7] Weiss Y, Torralba A, Fergus R. Spectral hashing[C]//Advances in Neural Information Processing Systems. British Columbia, Canada, 2008:1-8.
[8] Lazebnik S, Raginsky M. Locality-sensitive binary codes from shift-invariant kernels [C]//Advances in Neural Information Processing Systems. British Columbia, Canada, 2009: 1509-1517.
[9] Salakhutdinov R, Hinton G. Semantic hashing[J]. International Journal of Approximate Reasoning, 2009, 50(7): 969-978.
[10] Kulis B, Darrell T. Learning to hash with binary reconstructive embeddings[C]//Advances in Neural Information Processing Systems. Vancouver, Canada, 2009:1042-1050.
[11] Norouzi M, Fleet D. Minimal loss hashing for compact binary codes[C]//Proceedings of the 28th International Conference on Machine Learning. Bellevue, WA, USA, 2011: 353-360.
[12] Wang J, Kumar S, Chang S F. Semi-supervised hashing for large-scale search[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(12):2393-2406.
[13] Liu W, Wang J, Kumar S, et al. Hashing with graphs[C]//Proceedings of the 28th International Conference on Machine Learning. Bellevue, WA, USA, 2011: 1-8.
[14] Strecha C, Bronstein, A M, Bronstein M M, et al. LDAHash: improved matching with smaller descriptors[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(1):66-78.
[15] Andoni A, Indyk P. Near-optimal hashing algorithms for approximate nearest neighbor in high dimensions[C]//47th Annual IEEE Symposium on Foundations of Computer Science. Berkeley, CA, USA, 2006:459-468.
[16] Datar M, Immorlica N, Indyk P, et al. Locality-sensitive hashing scheme based on p-stable distributions[C]//Proceedings of the 20th Annual Symposium on Computational Geometry. New York, USA, 2004:253-262.
[17] Shakhnarovich G, Viola P, Darrell T. Fast pose estimation with parameter-sensitive hashing[C]//Proceedings of the 9th IEEE International Conference on Computer Vision. Nice, France, 2003: 750-757.
[18] Shen J F, Sun C Y, Yang W K, et al. A novel distribution-based feature for rapid object detection[J]. Neurocomputing, 2011, 74(17):2767-2779.
[19] Chang H, Ai H Z, Li Y, et al. High-performance rotation invariant multiview face detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(4): 671-686.

Memo

Memo:
Biographies: Zuo Xin(1980—), female, graduate; Luo Limin(corresponding author), male, doctor, professor, luo.list@seu.edu.cn.
Foundation items: The National Natural Science Foundation of China(No.61305058), the Natural Science Foundation of Higher Education Institutions of Jiangsu Province(No.12KJB520003), the Natural Science Foundation of Jiangsu Province(No.BK20130471), the Scientific Research Foundation for Advanced Talents by Jiangsu University(No.13JDG093).
Citation: Zuo Xin, Luo Limin, Shen Jifeng, et al. Learning compact binary code based on multiple heterogeneous features.[J].Journal of Southeast University(English Edition), 2013, 29(4):372-378.[doi:10.3969/j.issn.1003-7985.2013.04.004]
Last Update: 2013-12-20