Similar articles

Abstract: With the rapid development of the Internet, recent years have seen the explosive growth of social media. This brings great challenges in performing efficient and accurate image retrieval on a large scale. Recent work shows that using hashing methods to embed high-dimensional image features and tag information into Hamming space provides a powerful way to index large collections of social images. By learning hash codes through a spectral graph partitioning algorithm, spectral hashing (SH) has shown promising performance among various hashing approaches. However, it is incomplete to model the relations among images only by pairwise simple graphs which ignore the relationship in a higher order. In this paper, we utilize a probabilistic hypergraph model to learn hash codes for social image retrieval. A probabilistic hypergraph model offers a higher order representation among social images by connecting more than two images in one hyperedge. Unlike a normal hypergraph model, a probabilistic hypergraph model considers not only the grouping information, but also the similarities between vertices in hyperedges. Experiments on Flickr image datasets verify the performance of our proposed approach.

基于概率超图哈希编码的社交图像检索研究

研究目的：在过去十多年中，互联网多媒体数据爆炸性增长，数以亿计的网络图片给传统图像检索技术带来了巨大挑战。在如此庞大数据量上进行特征空间欧式距离最近邻搜索，不切实际。如何更有效地表达和检索网络图片成为当前研究热点。针对此热点和难点，本文提出了基于概率超图哈希编码的大规模图像快速检索技术。
创新要点：利用概率超图建立社交图片之间语义层面和视觉特征层面的关联性。相比简单图模型，超图模型能更有效地描述不同图片之间的高层次联系，寻找社交网络图片之间更深层次的信息。相比一般超图，概率超图能更有效地表述节点对超边的归属程度。利用超图拉普拉斯矩阵将概率超图投影到汉明（Hamming）空间，极大提升了图像存储、检索效率。
方法提亮：本方法结合了社交网络图片的视觉特征和用户标注信息，利用概率超图挖掘这两种信息的高层次关联性，并根据具体情况给予这两种信息不同权重。
重要结论：实验数据表明，与现有哈希检索方法相比，该方法对社交图像进行快速检索的准确率有较大提升。
超图拉普拉斯；概率超图；哈希编码；图像检索

[1]Andoni, A., Indyk, P., 2006. Near-optimal hashing algorithms for approximate nearest neighbor in high dimensions. 47th Annual IEEE Symp. on Foundations of Computer Science, p.459-468.

[2]Arya, S., Mount, D.M., Netanyahu, N.S., et al., 1998. An optimal algorithm for approximate nearest neighbor searching fixed dimensions. JACM, 45(6):891-923.

[3]Bengio, Y., Delalleau, O., le Roux, N., et al., 2004. Learning eigenfunctions links spectral embedding and kernel PCA. Neur. Comput., 16(10):2197-2219.

[4]Chua, T.S., Tang, J., Hong, R., et al., 2009. NUS-WIDE: a real-world web image database from National University of Singapore. Proc. ACM Int. Conf. on Image and Video Retrieval, p.48.

[5]Gao, Y., Wang, M., Zha, Z.J., et al., 2013. Visual-textual joint relevance learning for tag-based social image search. IEEE Trans. Image Process., 22(1):363-376.

[6]He, J., Li, M., Zhang, H.J., et al., 2004. Manifold-ranking based image retrieval. Proc. 12th Annual ACM Int. Conf. on Multimedia, p.9-16.

[7]He, J., Li, M., Zhang, H.J., et al., 2006. Generalized manifold-ranking-based image retrieval. IEEE Trans. Image Process., 15(10):3170-3177.

[8]Hinton, G.E., Salakhutdinov, R.R., 2006. Reducing the dimensionality of data with neural networks. Science, 313(5786):504-507.

[9]Huang, Y., Liu, Q., Zhang, S., et al., 2010. Image retrieval via probabilistic hypergraph ranking. IEEE Conf. on Computer Vision and Pattern Recognition, p.3376-3383.

[10]Jiang, Y.G., Ngo, C.W., 2008. Bag-of-visual-words expansion using visual relatedness for video indexing. Proc. 31st Annual Int. ACM SIGIR Conf. on Research and Development in Information Retrieval, p.769-770.

[11]Li, P., Wang, M., Cheng, J., et al., 2013. Spectral hashing with semantically consistent graph for image indexing. IEEE Trans. Multimedia, 15(1):141-152.

[12]Liu, H., le Pendu, P., Jin, R., et al., 2011. A hypergraph-based method for discovering semantically associated itemsets. IEEE 11th Int. Conf. on Data Mining, p.398-406.

[13]Lowe, D.G., 2004. Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis., 60(2):91-110.

[14]Salakhutdinov, R., Hinton, G.E., 2007. Learning a nonlinear embedding by preserving class neighborhood structure. Int. Conf. on Artificial Intelligence and Statistics, p.412-419.

[15]Shakhnarovich, G., Viola, P., Darrel, T., 2003. Fast pose estimation with parameter-sensitive hashing. Proc. 9th IEEE Int. Conf. on Computer Vision, p.750-757.

[16]Silpa-Anan, C., Hartley, R., 2008. Optimised KD-trees for fast image descriptor matching. IEEE Conf. on Computer Vision and Pattern Recognition, p.1-8.

[17]Torralba, A., Fergus, R., Weiss, Y., 2008. Small codes and large image databases for recognition. IEEE Conf. on Computer Vision and Pattern Recognition, p.1-8.

[18]Weiss, Y., Torralba, A., Fergus, R., 2008. Spectral hashing. 21st Advances in NIPS, p.1753-1760.

[19]Xia, S., Hancock, E.R., 2008. Clustering using class specific hyper graphs. Structural, Syntactic, and Statistical Pattern Recognition, p.318-328.

[20]Yang, J., Jiang, Y.G., Hauptmann, A.G., et al., 2007. Evaluating bag-of-visual-words representations in scene classification. Proc. Int. Workshop on Multimedia Information Retrieval, p.197-206.

[21]Yu, J., Tao, D., Wang, M., 2012. Adaptive hypergraph learning and its application in image classification. IEEE Trans. Image Process., 21(7):3262-3272.

[22]Zass, R., Shashua, A., 2008. Probabilistic graph and hypergraph matching. IEEE Conf. on Computer Vision and Pattern Recognition, p.1-8.

[23]Zhang, D., Wang, J., Cai, D., et al., 2010. Self-taught hashing for fast similarity search. Proc. 33rd Int. ACM SIGIR Conf. on Research and Development in Information Retrieval, p.18-25.

[24]Zhou, D., Bousquet, O., Lal, T.N., et al., 2004. Learning with local and global consistency. 17th Advances in NIPS, p.321-328.

[25]Zhou, D., Huang, J., Schölkopf, B., 2006. Learning with hypergraphs: clustering, classification, and embedding. 19th Advances in NIPS, p.1601-1608.

[26]Zhuang, Y., Liu, Y., Wu, F., et al., 2011. Hypergraph spectral hashing for similarity search of social image. Proc. 19th ACM Int. Conf. on Multimedia, p.1457-1460.