Performing content-based retrieval of humans using gait biometrics

In order to analyse surveillance video, we need to efficiently explore large datasets containing videos of walking humans. Effective analysis of such data relies on retrieval of video data which has been enriched using semantic annotations. A manual annotation process is time-consuming and prone to error due to subject bias however, at surveillance-image resolution, the human walk (their gait) can be analysed automatically. We explore the content-based retrieval of videos containing walking subjects, using semantic queries. We evaluate current research in gait biometrics, unique in its effectiveness at recognising people at a distance. We introduce a set of semantic traits discernible by humans at a distance, outlining their psychological validity. Working under the premise that similarity of the chosen gait signature implies similarity of certain semantic traits we perform a set of semantic retrieval experiments using popular Latent Semantic Analysis techniques. We perform experiments on a dataset of 2000 videos of people walking in laboratory conditions and achieve promising retrieval results for features such as Sex (mAP  =  14% above random), Age (mAP  =  10% above random) and Ethnicity (mAP  =  9% above random).     

基于多哈希算法的大规模图像快速检索方法

哈希技术被视为最有潜力的相似性搜索方法,其可以用于大规模多媒体数据搜索场合。为了解决在大规模图像情况下,数据检索效率低下的问题,提出了一种基于分段哈希码的倒排索引树结构,该索引结构将哈希码进行分段处理,对每段哈希码维护一个倒排索引树结构,并结合高效的布隆过滤器构建哈希索引结构。为了进一步提高检索准确性,设计了一种准确的排序融合算法,对多个哈希算法的排序结果分别构建加权无向图,采用PageRank的思想对基于多个哈希算法的排序列表的融合技术进行了详细的说明。实验结果表明,基于分段哈希码的倒排索引树结构能极大地提升数据的检索速度。此外,相比于传统的单个哈希算法排序技术,基于多个哈希算法的排序列表融合技术的检索准确率优势显著。     

Efficient kNN search in polyphonic music databases using a lower bounding mechanism

Querying polyphonic music from a large data collection is an interesting topic. Recently, researchers have attempted to provide efficient methods for content-based retrieval in polyphonic music databases where queries are polyphonic. However, most of them do not work well for similarity search, which is important to many applications. In this paper, we propose three polyphonic representations with the associated similarity measures and a novel method to retrieve k music works that contain segments most similar to the query. In general, most of the index-based methods for similarity search generate all the possible answers to the query and then perform exact matching on the index for each possible answer. Based on the edit distance, our method generates only a few possible answers by performing the deletion and/or replacement operations on the query. Each possible answer is then used to perform exact matching on a list-based index, which allows the insertion operations to be performed. For each possible answer, its edit distance to the query is regarded as a lower bound of the edit distances between the matched results and the query. Based on the kNN results that match a possible answer, the possible answers that cannot provide better results are skipped. By using this mechanism, we design a method for efficient kNN search in polyphonic music databases. The experimental results show that our method outperforms the previous methods in efficiency. We also evaluate the effectiveness of our method by showing the search results to the musician and nonmusician user groups. The experimental results provide useful guidelines on the design of a polyphonic music database.     

Classification and function prediction of proteins using data compression

Protein has a complicated spatial structure, and has chemical and physical functions which originate from this structure. It is important to predict the structure and function of proteins from a DNA sequence or amino acid sequence from the viewpoint of biology, medical science, protein engineering, etc. However, to data there is no way to predict them accurately from these sequences. Instead, some approaches attempt to estimate the functions based on an approximate similarity in the retrieval of sequences. We propose a new method for the similarity retrieval of an amino acid sequence based on the concept of homology retrieval using data compression. The introduction of compression by a dictionary technique enables us to describe the text data as ann-dimensional vector usingn dictionaries, which is generated by compressingn typical texts, and enables us to classify the proteins based on their similarity. We examined the effectiveness of our proposal using real genome data. This work was presented in part at the Sixth International Symposium on Artificial Life and Robotics, Tokyo, January 15–17, 2001     

Exploiting geographic references of documents in a geographical information retrieval system using an ontology-based index

Both Geographic Information Systems and Information Retrieval have been very active research fields in the last decades. Lately, a new research field called Geographic Information Retrieval has appeared from the intersection of these two fields. The main goal of this field is to define index structures and techniques to efficiently store and retrieve documents using both the text and the geographic references contained within the text. We present in this paper two contributions to this research field. First, we propose a new index structure that combines an inverted index and a spatial index based on an ontology of geographic space. This structure improves the query capabilities of other proposals. Then, we describe the architecture of a system for geographic information retrieval that defines a workflow for the extraction of the geographic references in documents. The architecture also uses the index structure that we propose to solve pure spatial and textual queries as well as hybrid queries that combine both a textual and a spatial component. Furthermore, query expansion can be performed on geographic references because the index structure is based in an ontology.     

The <Emphasis Type="Italic">k</Emphasis>-Nearest Neighbour Join: Turbo Charging the KDD Process

The similarity join has become an important database primitive for supporting similarity searches and data mining. A similarity join combines two sets of complex objects such that the result contains all pairs of similar objects. Two types of the similarity join are well-known, the distance range join, in which the user defines a distance threshold for the join, and the closest pair query or k-distance join, which retrieves the k most similar pairs. In this paper, we propose an important, third similarity join operation called the k-nearest neighbour join, which combines each point of one point set with its k nearest neighbours in the other set. We discover that many standard algorithms of Knowledge Discovery in Databases (KDD) such as k-means and k-medoid clustering, nearest neighbour classification, data cleansing, postprocessing of sampling-based data mining, etc. can be implemented on top of the k-nn join operation to achieve performance improvements without affecting the quality of the result of these algorithms. We propose a new algorithm to compute the k-nearest neighbour join using the multipage index (MuX), a specialised index structure for the similarity join. To reduce both CPU and I/O costs, we develop optimal loading and processing strategies.     

Discovery of a perceptual distance function for measuring image similarity

For more than a decade, researchers have actively explored the area of image/video analysis and retrieval. Yet one fundamental problem remains largely unsolved: how to measure perceptual similarity between two objects. For this purpose, most researchers employ a Minkowski-type metric. Unfortunately, the Minkowski metric does not reliably find similarities in objects that are obviously alike. Through mining a large set of visual data, our team has discovered a perceptual distance function. We call the discovered function the dynamic partial function (DPF). When we empirically compare DPF to Minkowski-type distance functions in image retrieval and in video shot-transition detection using our image features, DPF performs significantly better. The effectiveness of DPF can be explained by similarity theories in cognitive psychology.     

Indexing expensive functions for efficient multi-dimensional similarity search

Similarity search is important in information retrieval applications where objects are usually represented as vectors of high dimensionality. This leads to the increasing need for supporting the indexing of high-dimensional data. On the other hand, indexing structures based on space partitioning are powerless because of the well-known “curse of dimensionality”. Linear scan of the data with approximation is more efficient in the high-dimensional similarity search. However, approaches so far have concentrated on reducing I/O, and ignored the computation cost. For an expensive distance function such as L _p norm with fractional p, the computation cost becomes the bottleneck. We propose a new technique to address expensive distance functions by “indexing the function” by pre-computing some key values of the function once. Then, the values are used to develop the upper/lower bounds of the distance between a data vector and the query vector. The technique is extremely efficient since it avoids most of the distance function computations; moreover, it does not involve any extra secondary storage because no index is constructed and stored. The efficiency is confirmed by cost analysis, as well as experiments on synthetic and real data.     

Indexing shapes in image databases using the centroid–radii model

In content-based image retrieval systems, the content of an image such as color, shapes and textures are used to retrieve images that are similar to a query image. Most of the existing work focus on the retrieval effectiveness of using content for retrieval, i.e., study the accuracy (in terms of recall and precision) of using different representations of content. In this paper, we address the issue of retrieval efficiency, i.e., study the speed of retrieval, since a slow system is not useful for large image databases. In particular, we look at using the shape feature as the content of an image, and employ the centroid–radii model to represent the shape feature of objects in an image. This facilitates multi-resolution and similarity retrievals. Furthermore, using the model, the shape of an object can be transformed into a point in a high-dimensional data space. We can thus employ any existing high-dimensional point index as an index to speed up the retrieval of images. We propose a multi-level R-tree index, called the Nested R-trees (NR-trees) and compare its performance with that of the R-tree. Our experimental study shows that NR-trees can reduce the retrieval time significantly compared to R-tree, and facilitate similarity retrieval. We note that our NR-trees can also be used to index high-dimensional point data commonly found in many other applications.     

一种支持快速相似检索的多维索引结构

基于内容的图像检索是一种典型的相似检索问题,对于尺度空间上的图像相似匹配问题,一般认为距离计算费用很高.因此,需要建立有效的索引结构,以减少每个查询中的距离计算次数.为此,基于数据空间的"优化划分",并且使用"代表点",以层次结构方式划分数据,提出了一种新的基于距离的相似索引结构opt-树及其变种(-树.为了更有效地支持基于内容的图像检索,在(-树索引结构中采用了"(-最优化划分"和"(-对称冗余存储"策略,以提高相似检索的效率.详细讨论了这种索引结构的建立与检索等问题,并给出了相应的算法.实验结果显示了这种索引技术的有效性.     

Range query processing on single and multi GPU environments

Metric-space similarity search has proven suitable in a number of application domains such as multimedia retrieval and computational biology to name a few. These applications usually work on very large databases that are often indexed to speed-up on-line searches. To achieve efficient throughput, it is essential to exploit the intrinsic parallelism in the respective search query processing algorithms. Many strategies have been proposed in the literature to parallelize these algorithms either on shared or distributed memory multiprocessor systems. Lately, GPUs have been used to implement brute-force parallel search strategies instead of using index data structures. Indexing poses difficulties when it comes to achieve efficient exploitation of GPU resources. In this paper we propose single and multi GPU metric space techniques that efficiently exploit GPU tailored index data structures for parallel similarity search in large databases. The experimental results show that our proposal outperforms previous index-based sequential and OpenMP search strategies.     

应用多索引加法量化编码的近邻检索算法

目的 海量图像检索技术是计算机视觉领域研究热点之一,一个基本的思路是对数据库中所有图像提取特征,然后定义特征相似性度量,进行近邻检索。海量图像检索技术,关键的是设计满足存储需求和效率的近邻检索算法。为了提高图像视觉特征的近似表示精度和降低图像视觉特征的存储空间需求,提出了一种多索引加法量化方法。方法 由于线性搜索算法复杂度高,而且为了满足检索的实时性,需把图像描述符存储在内存中,不能满足大规模检索系统的需求。基于非线性检索的优越性,本文对非穷尽搜索的多索引结构和量化编码进行了探索新研究。利用多索引结构将原始数据空间划分成多个子空间,把每个子空间数据项分配到不同的倒排列表中,然后使用压缩编码的加法量化方法编码倒排列表中的残差数据项,进一步减少对原始空间的量化损失。在近邻检索时采用非穷尽搜索的策略,只在少数倒排列表中检索近邻项,可以大大减少检索时间成本,而且检索过程中不用存储原始数据,只需存储数据集中每个数据项在加法量化码书中的码字索引,大大减少内存消耗。结果 为了验证算法的有效性,在3个数据集SIFT、GIST、MNIST上进行测试,召回率相比近几年算法提升4%~15%,平均查准率提高12%左右,检索时间与最快的算法持平。结论 本文提出的多索引加法量化编码算法,有效改善了图像视觉特征的近似表示精度和存储空间需求,并提升了在大规模数据集的检索准确率和召回率。本文算法主要针对特征进行近邻检索,适用于海量图像以及其他多媒体数据的近邻检索。     

Finding approximate matches in large lexicons

Approximate string matching is used for spelling correction and personal name matching. In this paper we show how to use string matching techniques in conjunction with lexicon indexes to find approximate matches in a large lexicon. We test several lexicon indexing techniques, including n-grams and permuted lexicons, and several string matching techniques, including string similarity measures and phonetic coding. We propose methods for combining these techniques, and show experimentally that these combinations yield good retrieval effectiveness while keeping index size and retrieval time low. Our experiments also suggest that, in contrast to previous claims, phonetic codings are markedly inferior to string distance measures, which are demonstrated to be suitable for both spelling correction and personal name matching.     

MINDEX: An efficient index structure for salient-object-based queries in video databases

Several salient-object-based data models have been proposed to model video data. However, none of them addresses the development of an index structure to efficiently handle salient-object-based queries. There are several indexing schemes that have been proposed for spatiotemporal relationships among objects, and they are used to optimize timestamp and interval queries, which are rarely used in video databases. Moreover, these index structures are designed without consideration of the granularity levels of constraints on salient objects and the characteristics of video data. In this paper, we propose a multilevel index structure (MINDEX) to efficiently handle the salient-object-based queries with different levels of constraints. We present experimental results showing the performance of different methods of MINDEX construction.     

VAR-Tree--一种新的高维数据索引结构

在多媒体信息检索和数据挖掘等应用领域,实现高维矢量的K近邻搜索是非常具有挑战性的研究课题,为此人们提出了很多种索引结构．然而,现有研究成果表明,随着矢量维数的增加,基于树状索引结构的查询性能急剧下降,例如在R-Tree,X-Tree和SS-Tree中都会出现“维数灾难”．为此,又引入近似压缩的思想,即通过压缩数据来减少查询过程中的磁盘读写代价,例如VA-File等,不过,VA-File没有对近似矢量数据做任何的排序或层次处理．提出了一种新的索引结构VAR-Tree,它将VA-File与R-Tree有机结合起来,用R-Tree管理和组织VA-File中的近似数据,并用已提出的R-Tree类相似查询算法实现基于VAR-Tree的查询．实验结果表明,VAR-Tree较好地提高了检索性能．     

Efficient Similarity Search in Nonmetric Spaces with Local Constant Embedding

Similarity-based search has been a key factor for many applications such as multimedia retrieval, data mining, Web search and retrieval, and so on. There are two important issues related to the similarity search, namely, the design of a distance function to measure the similarity and improving the search efficiency. Many distance functions have been proposed, which attempt to closely mimic human recognition. Unfortunately, some of these well-designed distance functions do not follow the triangle inequality and are therefore nonmetric. As a consequence, efficient retrieval by using these nonmetric distance functions becomes more challenging, since most existing index structures assume that the indexed distance functions are metric. In this paper, we address this challenging problem by proposing an efficient method, that is, local constant embedding (LCE), which divides the data set into disjoint groups so that the triangle inequality holds within each group by constant shifting. Furthermore, we design a pivot selection approach for the converted metric distance and create an index structure to speed up the retrieval efficiency. Moreover, we also propose a novel method to answer approximate similarity search in the nonmetric space with a guaranteed query accuracy. Extensive experiments show that our method works well on various nonmetric distance functions and improves the retrieval efficiency by an order of magnitude compared to the linear scan and existing retrieval approaches with no false dismissals.     

A parallel computational approach for similarity search using Bloom filters

Finding similar items in a large and unstructured dataset is a challenging task in many applications of data science, such as searching, indexing, and retrieval. With the increasing data volume and demand for real time responses, similarity search has gained much consideration. In this paper, a parallel computational approach for similarity search using Bloom filters (PCASSB) has been proposed, which uses Bloom filter for the representation of features of document and comparison with user's query. Query features are stored in integer query array (^IQ_A), an array of integer. The PCASSB, an approximate similarity search technique, has been implemented on graphics processing unit with compute unified device architecture as the programming platform. To compute the similarity score between query and reference dataset, Dice coefficient has been used as a baseline method. The accuracy of the results generated by PCASSB is compared with the baseline method and other state‐of‐the‐art methods. The experimental results show that the proposed technique is quite effective in processing large number of text documents as it takes less computational time.     

The PN-Tree: A Parallel and Distributed Multidimensional Index

Multidimensional indexing is concerned with the indexing of multi-attributed records, where queries can be applied on some or all of the attributes. Indexing multi-attributed records is referred to by the term multidimensional indexing because each record is viewed as a point in a multidimensional space with a number of dimensions that is equal to the number of attributes. The values of the point coordinates along each dimension are equivalent to the values of the corresponding attributes. In this paper, the PN-tree, a new index structure for multidimensional spaces, is presented. This index structure is an efficient structure for indexing multidimensional points and is parallel by nature. Moreover, the proposed index structure does not lose its efficiency if it is serially processed or if it is processed using a small number of processors. The PN-tree can take advantage of as many processors as the dimensionality of the space. The PN-tree makes use of B⁺-trees that have been developed and tested over years in many DBMSs. The PN-tree is compared to the Hybrid tree that is known for its superiority among various index structures. Experimental results show that parallel processing of the PN-tree reduces significantly the number of disk accesses involved in the search operation. Even in its serial case, the PN-tree outperforms the Hybrid tree for large database sizes.