面向视觉搜索的空间局部敏感哈希方法

目的 视觉检索需要准确、高效地从大型图像或者视频数据集中检索出最相关的视觉内容,但是由于数据集中图像数据量大、特征维度高的特点,现有方法很难同时保证快速的检索速度和较好的检索效果。方法 对于面向图像视频数据的高维数据视觉检索任务,提出加权语义局部敏感哈希算法（weighted semantic locality-sensitive hashing, WSLSH）。该算法利用两层视觉词典对参考特征空间进行二次空间划分,在每个子空间里使用加权语义局部敏感哈希对特征进行精确索引。其次,设计动态变长哈希码,在保证检索性能的基础上减少哈希表数量。此外,针对局部敏感哈希（locality sensitive hashing, LSH）的随机不稳定性,在LSH函数中加入反映参考特征空间语义的统计性数据,设计了一个简单投影语义哈希函数以确保算法检索性能的稳定性。结果 在Holidays、Oxford5k和DataSetB数据集上的实验表明,WSLSH在DataSetB上取得最短平均检索时间0.034 25 s;在编码长度为64位的情况下,WSLSH算法在3个数据集上的平均精确度均值（mean average precision,mAP）分别提高了1.2%32.6%、1.7%19.1%和2.6%28.6%,与几种较新的无监督哈希方法相比有一定的优势。结论 通过进行二次空间划分、对参考特征的哈希索引次数进行加权、动态使用变长哈希码以及提出简单投影语义哈希函数来对LSH算法进行改进。由此提出的加权语义局部敏感哈希（WSLSH）算法相比现有工作有更快的检索速度,同时,在长编码的情况下,取得了更为优异的性能。     

基于柯西分布的深度哈希跨媒体检索

针对深度哈希跨媒体检索方法中,语义相似的媒体对象的哈希码在汉明空间内的分布不合理问题,提出了一种新的深度哈希跨媒体检索模型.该模型是在汉明空间内利用柯西分布对现有的深度哈希跨媒体关联损失进行改进,使得语义相似的媒体对象哈希码距离较小,语义不相似的媒体对象哈希码较大,进而提高模型的检索效果.同时,本文给出了一种高效的模型求解方法,采用交替迭代方式获得模型的近似最优解.在Flickr-25k数据集,IAPR TC-12数据集和MS COCO数据集上的实验结果表明,该方法可以有效的提高跨媒体检索性能.     

A Novel Approach Towards Large Scale Cross-Media Retrieval

With the rapid development of Internet and multimedia technology,cross-media retrieval is concerned to retrieve all the related media objects with multi-modality by submitting a query media object.Unfortunately,the complexity and the heterogeneity of multi-modality have posed the following two major challenges for cross-media retrieval:1) how to construct a unified and compact model for media objects with multi-modality,2) how to improve the performance of retrieval for large scale cross-media database.In this paper,we propose a novel method which is dedicate to solving these issues to achieve effective and accurate cross-media retrieval.Firstly,a multi-modality semantic relationship graph(MSRG) is constructed using the semantic correlation amongst the media objects with multi-modality.Secondly,all the media objects in MSRG are mapped onto an isomorphic semantic space.Further,an efficient indexing MK-tree based on heterogeneous data distribution is proposed to manage the media objects within the semantic space and improve the performance of cross-media retrieval.Extensive experiments on real large scale cross-media datasets indicate that our proposal dramatically improves the accuracy and efficiency of cross-media retrieval,outperforming the existing methods significantly.     

Cross-media retrieval using query dependent search methods

The content-based cross-media retrieval is a new type of multimedia retrieval in which the media types of query examples and the returned results can be different. In order to learn the semantic correlations among multimedia objects of different modalities, the heterogeneous multimedia objects are analyzed in the form of multimedia document (MMD), which is a set of multimedia objects that are of different media types but carry the same semantics. We first construct an MMD semi-semantic graph (MMDSSG) by jointly analyzing the heterogeneous multimedia data. After that, cross-media indexing space (CMIS) is constructed. For each query, the optimal dimension of CMIS is automatically determined and the cross-media retrieval is performed on a per-query basis. By doing this, the most appropriate retrieval approach for each query is selected, i.e. different search methods are used for different queries. The query dependent search methods make cross-media retrieval performance not only accurate but also stable. We also propose different learning methods of relevance feedback (RF) to improve the performance. Experiment is encouraging and validates the proposed methods.     

区块链环境中基于局部敏感哈希的协同过滤推荐研究

针对区块链环境中海量高维的数据使得推荐性能低下的问题,通过对局部敏感哈希算法的优化,降低其在近邻搜索过程中带来的额外计算和存储开销.利用数据分布的主成分减少传统LSH中不良捕获的投影方向,同时对投影向量权重进行量化,以减少哈希表和哈希函数的使用;通过对哈希桶的间隔进行调整,并且根据冲突次数的大小进一步细化查询结果集,以...     

Mining Semantic Correlation of Heterogeneous Multimedia Data for Cross-Media Retrieval

Although multimedia objects such as images, audios and texts are of different modalities, there are a great amount of semantic correlations among them. In this paper, we propose a method of transductive learning to mine the semantic correlations among media objects of different modalities so that to achieve the cross-media retrieval. Cross-media retrieval is a new kind of searching technology by which the query examples and the returned results can be of different modalities, e.g., to query images by an example of audio. First, according to the media objects features and their co-existence information, we construct a uniform cross-media correlation graph, in which media objects of different modalities are represented uniformly. To perform the cross-media retrieval, a positive score is assigned to the query example; the score spreads along the graph and media objects of target modality or MMDs with the highest scores are returned. To boost the retrieval performance, we also propose different approaches of long-term and short-term relevance feedback to mine the information contained in the positive and negative examples.     

基于可扩展LSH的高维动态数据索引

提出了一种可扩展的局部敏感哈希索引(SLSH),以解决高维动态数据索引中,由于数据集大小及分布特征无法确定而导致索引效率降低的问题.SLSH架构于E2LSH之上,继承了其对高维数据索引速度快,并可直接对欧式空间上的数据点进行索引的特点.为了使得哈希索引具有动态的相似性区分能力,SLSH修改了E2LSH的哈希族,通过哈希桶容量约束自适应调节哈希参数.因此对于分布密度动态变化的数据空间,SLSH也能够给出鲁棒的划分.     

两层传感网隐私保护的不完全数据Skyline查询协议

感知节点感知数据易受外界环境影响,使得不完全数据广泛存在于无线传感器网络中,且感知数据面临严重的隐私威胁。针对两层传感器网络不完全数据查询过程中存在的隐私泄露问题,提出一种基于置换和桶技术的两层传感器网络隐私保护的不完全数据Skyline查询协议（PPIS）。为了实现对不完全数据的Skyline查询,PPIS将缺失属性值置换为数据域的上界值,并将不完全数据映射到桶中;为了保证数据隐私性,PPIS首先将桶区间转化为前缀编码,然后将前缀编码加载到Bloom过滤器中,保证存储节点在无需数据和桶区间明文的前提下执行查询处理;为了保证查询结果的完整性,PPIS采用Merkle哈希树构造完整性验证编码,实现对查询结果的完整性验证。理论分析和仿真实验验证了PPIS的安全性和有效性,与现有隐私保护Skyline查询协议SMQ和SSQ相比,PPIS通信能耗节省了70%以上。     

Locality preserving hashing for fast image search: theory and applications

Abstract

State-of-the-art hashing methods, such as the kernelised locality-sensitive hashing and spectral hashing, have high algorithmic complexities to build the hash codes and tables. Our observation from the existing hashing method is that, putting two dissimilar data points into the same hash bucket only reduces the efficiency of the hash table, but it does not hurt the query accuracy. Whereas putting two similar data points into different hash buckets will reduce the correctness (i.e. query accuracy) of a hashing method. Therefore, it is much more important for a good hashing method to ensure that similar data points have high probabilities to be put to the same bucket, than considering those dissimilar data-point relations. On the other side, attracting similar data points to the same hash bucket will naturally suppress dissimilar data points to be put into the same hash bucket. With this locality-preserving observation, we naturally propose a new hashing method called the locality-preserving hashing, which builds the hash codes and tables with much lower algorithmic complexity. Experimental results show that the proposed method is very competitive in terms of the training time spent for large data-sets among the state of the arts, and with reasonable or even better query accuracy.     

Semi-paired and semi-supervised multimodal hashing via cross-modality label propagation

Multimedia Tools and Applications - Due to the fast query speed and low storage cost, multimodal hashing methods have been attracting increasing attention in large-scale cross-media retrieval...     

Harmonizing Hierarchical Manifolds for Multimedia Document Semantics Understanding and Cross-Media Retrieval

In this paper, we consider the problem of multimedia document (MMD) semantics understanding and content-based cross-media retrieval. An MMD is a set of media objects of different modalities but carrying the same semantics and the content-based cross-media retrieval is a new kind of retrieval method by which the query examples and search results can be of different modalities. Two levels of manifolds are learned to explore the relationships among all the data in the level of MMD and in the level of media object respectively. We first construct a Laplacian media object space for media object representation of each modality and an MMD semantic graph to learn the MMD semantic correlations. The characteristics of media objects propagate along the MMD semantic graph and an MMD semantic space is constructed to perform cross-media retrieval. Different methods are proposed to utilize relevance feedback and experiment shows that the proposed approaches are effective.     

深度多模态融合服装风格检索

目的 服装检索方法是计算机视觉与自然语言处理领域的研究热点,其包含基于内容与基于文本的两种查询模态。然而传统检索方法通常存在检索效率低的问题,且很少研究关注服装在风格上的相似性。为解决这些问题,本文提出深度多模态融合的服装风格检索方法。方法 提出分层深度哈希检索模型,基于预训练的残差网络ResNet（residual network）进行迁移学习,并把分类层改造成哈希编码层,利用哈希特征进行粗检索,再用图像深层特征进行细检索。设计文本分类语义检索模型,基于LSTM（long short-term memory）设计文本分类网络以提前分类缩小检索范围,再以基于doc2vec提取的文本嵌入语义特征进行检索。同时提出相似风格上下文检索模型,其参考单词相似性来衡量服装风格相似性。最后采用概率驱动的方法量化风格相似性,并以最大化该相似性的结果融合方法作为本文检索方法的最终反馈。结果 在Polyvore数据集上,与原始ResNet模型相比,分层深度哈希检索模型的top5平均检索精度提高11.6%,检索速度提高2.57 s/次。与传统文本分类嵌入模型相比,本文分类语义检索模型的top5查准率提高29.96%,检索速度提高16.53 s/次。结论 提出的深度多模态融合的服装风格检索方法获得检索精度与检索速度的提升,同时进行了相似风格服装的检索使结果更具有多样性。     

Token list based information search in a multi-dimensional massive database

Finding proximity information is crucial for massive database search. Locality Sensitive Hashing (LSH) is a method for finding nearest neighbors of a query point in a high-dimensional space. It classifies high-dimensional data according to data similarity. However, the “curse of dimensionality” makes LSH insufficiently effective in finding similar data and insufficiently efficient in terms of memory resources and search delays. The contribution of this work is threefold. First, we study a Token List based information Search scheme (TLS) as an alternative to LSH. TLS builds a token list table containing all the unique tokens from the database, and clusters data records having the same token together in one group. Querying is conducted in a small number of groups of relevant data records instead of searching the entire database. Second, in order to decrease the searching time of the token list, we further propose the Optimized Token list based Search schemes (OTS) based on index-tree and hash table structures. An index-tree structure orders the tokens in the token list and constructs an index table based on the tokens. Searching the token list starts from the entry of the token list supplied by the index table. A hash table structure assigns a hash ID to each token. A query token can be directly located in the token list according to its hash ID. Third, since a single-token based method leads to high overhead in the results refinement given a required similarity, we further investigate how a Multi-Token List Search scheme (MTLS) improves the performance of database proximity search. We conducted experiments on the LSH-based searching scheme, TLS, OTS, and MTLS using a massive customer data integration database. The comparison experimental results show that TLS is more efficient than an LSH-based searching scheme, and OTS improves the search efficiency of TLS. Further, MTLS per forms better than TLS when the number of tokens is appropriately chosen, and a two-token adjacent token list achieves the shortest query delay in our testing dataset.     

移动点对象HR索引及反向最近邻查询

在此提出了一种基于速度分布的HR树索引结构,首先在速度域中对移动对象集进行规则划分,根据速度标量大小将移动对象划分到不同的速度树中,每棵速度树中移动对象具有相近的速度;对每棵速度树中的移动对象,则利用时间间隔进行划分。HR树索引增加了两个分别建于叶节点和根节点之上的Hash辅助索引结构,并基于HR树提出了反向最近邻查询算法,具有很好的动态更新性能和并发性。实验结果与分析表明,基于HR树索引的反向最近邻查询算法具有良好的更新及查询性能,优于通用的TPR树索引。     

Joint graph regularization based modality-dependent cross-media retrieval

Cross-media retrieval returns heterogeneous multimedia data of the same semantics for a query object, and the key problem for cross-media retrieval is how to deal with the correlations of heterogeneous multimedia data. Many works focus on mapping different modal data into an isomorphic space, so the similarities between different modal data can be measured. Inspired by this idea, we propose a joint graph regularization based modality-dependent cross-media retrieval approach (JGRMDCR), which takes into account the one-to-one correspondence between different modal data pairs, the inter-modality similarities and the intra-modality similarities. Meanwhile, according to the modality of the query object, this method learns different projection matrices for different retrieval tasks. Experimental results on benchmark datasets show that the proposed approach outperforms the other state-of-the-art algorithms.     

结合Transformer与非对称学习策略的图像检索

目的 图像检索是计算机视觉领域的一项基础任务,大多采用卷积神经网络和对称式学习策略,导致所需训练数据量大、模型训练时间长、监督信息利用不充分。针对上述问题,本文提出一种Transformer与非对称学习策略相结合的图像检索方法。方法 对于查询图像,使用Transformer生成图像的哈希表示,利用哈希损失学习哈希函数,使图像的哈希表示更加真实。对于待检索图像,采用非对称式学习策略,直接得到图像的哈希表示,并将哈希损失与分类损失相结合,充分利用监督信息,提高训练速度。在哈希空间通过计算汉明距离实现相似图像的快速检索。结果 在CIFAR-10和NUS-WIDE两个数据集上,将本文方法与主流的5种对称式方法和性能最优的两种非对称式方法进行比较,本文方法的mAP(mean average precision)比当前最优方法分别提升了5.06%和4.17%。结论 本文方法利用Transformer提取图像特征,并将哈希损失与分类损失相结合,在不增加训练数据量的前提下,减少了模型训练时间。所提方法性能优于当前同类方法,能够有效完成图像检索任务。     

目标提取与哈希机制的多标签图像检索

目的 哈希是大规模图像检索的有效方法。为提高检索精度,哈希码应保留语义信息。图像之间越相似,其哈希码也应越接近。现有方法首先提取描述图像整体的特征,然后生成哈希码。这种方法不能精确地描述图像包含的多个目标,限制了多标签图像检索的精度。为此提出一种基于卷积神经网络和目标提取的哈希生成方法。方法 首先提取图像中可能包含目标的一系列区域,然后用深度卷积神经网络提取每个区域的特征并进行融合,通过生成一组特征来刻画图像中的每个目标,最后再产生整幅图像的哈希码。采用Triplet Loss的训练方法,使得哈希码尽可能保留语义信息。结果 在VOC2012、Flickr25K和NUSWIDE数据集上进行多标签图像检索。在NDCG(normalized discounted cumulative gain)性能指标上,当返回图像数量为 1 000时,对于VOC2012,本文方法相对于DSRH(deep semantic ranking hashing)方法提高24个百分点,相对于ITQ-CCA(iterative quantization-canonical correlation analysis)方法能提高36个百分点;对于Flickr25,本文方法比DSRH方法能提高2个左右的百分点;对于NUSWIDE,本文方法相对于DSRH方法能提高4个左右的百分点。对于平均检索准确度,本文方法在NUSWIDE和Flickr25上能提高25个百分点。根据多项评价指标可以看出,本文方法能以更细粒度来精确地描述图像,显著提高了多标签图像检索的性能。结论 本文新的特征学习模型,对图像进行细粒度特征编码是一种可行的方法,能够有效提高数据集的检索性能。     

一种支持海量跨媒体检索的集成索引结构

提出一种支持海量跨媒体检索的集成索引结构.该方法首先通过对网页的预处理,分析其中不同模态媒体对象之间的链接关系,生成交叉参照图.然后通过用户相关反馈进行调节.当用户提交一个查询对象时,首先对交叉参照图进行基于索引的快速定位,得到与查询对象相关的候选媒体对象.然后对得到的候选媒体对象进行距离运算,得到结果媒体对象.理论分析和实验表明,该方法较顺序检索具有更好的性能,非常适合海量跨媒体数据检索.     

基于E2LSH的轨迹KNN查询算法

目前海量时空轨迹数据近邻查询算法中存在计算时间复杂度较高的问题,因此提出了一种结合领域POI数据和E2LSH算法的轨迹KNN查询算法。首先利用GeoHash技术对地理空间进行编码,然后结合POI数据实现向量空间的初步降维,进而根据停留时间构建每条轨迹的向量,采用局部敏感哈希函数运算结果建立轨迹索引,最后对查询返回的相似轨迹集合分别进行距离计算,经过排序得到距离最近的K个查询结果。对于增量的轨迹数据,利用E2LSH算法计算哈希值,直接添加轨迹索引,从而避免了复杂的计算过程以及对现有轨迹索引的影响。基于合成数据及真实数据集的实验结果表明,该方法在海量时空轨迹数据的近邻查询中,虽然牺牲了一定的准确率,但有效提升了算法效率,并能够高效简便地处理增量的时空轨迹数据。