Multiple representations, similarity matching, and results fusion for content-based image retrieval

In this paper, we show how the use of multiple content representations and their fusion can improve the performance of content-based image retrieval systems. We consider the case of texture and propose a new algorithm for texture retrieval based on multiple representations and their results fusion. Texture content is modeled using two different models: the well-known autoregressive model and a perceptual model based on perceptual features such as coarseness and directionality. In the case of the perceptual model, two viewpoints are considered: perceptual features are computed based on the original images viewpoint and on the autocovariance function viewpoint (corresponding to original images). So we consider a total of three content representations. The similarity measure used is based on Gower's index of similarity. Simple results of the fusion models are used to merge search results returned by different representations. Experimentations and benchmarking carried out on the well-known Brodatz database show a drastic improvement in search effectiveness with the fused model without necessarily altering their efficiency in an important way.     

Relevance feature mapping for content-based multimedia information retrieval

This paper presents a novel ranking framework for content-based multimedia information retrieval (CBMIR). The framework introduces relevance features and a new ranking scheme. Each relevance feature measures the relevance of an instance with respect to a profile of the targeted multimedia database. We show that the task of CBMIR can be done more effectively using the relevance features than the original features. Furthermore, additional performance gain is achieved by incorporating our new ranking scheme which modifies instance rankings based on the weighted average of relevance feature values. Experiments on image and music databases validate the efficacy and efficiency of the proposed framework.     

A survey on content-based retrieval for multimedia databases

Conventional database systems are designed for managing textual and numerical data, and retrieving such data is often based on simple comparisons of text/numerical values. However, this simple method of retrieval is no longer adequate for multimedia data, since the digitized representation of images, video, or data itself does not convey the reality of these media items. In addition, composite data consisting of heterogeneous types of data also associates with the semantic content acquired by a user's recognition. Therefore, content-based retrieval for multimedia data is realized taking such intrinsic features of multimedia data into account. Implementation of the content-based retrieval facility is not based on a single fundamental, but is closely related to an underlying data model, a priori knowledge of the area of interest, and the scheme for representing queries. This paper surveys recent studies on content-based retrieval for multimedia databases from the point of view of three fundamental issues. Throughout the discussion, we assume databases that manage only nontextual/numerical data, such as image or video, are also in the category of multimedia databases     

Multiple-instance content-based image retrieval employing isometric embedded similarity measure

In image-based retrieval, global or local features sufficiently discriminative to summarize the image content are commonly extracted first. Traditional features, such as color, texture, shape or corner, characterizing image content are not reliable in terms of similarity measure. A good match in the feature domain does not necessarily map to image pairs with similar relationship. Applying these features as search keys may retrieve dissimilar false-positive images, or leave similar false-negative ones behind. Moreover, images are inherently ambiguous since they contain a great amount of information that justifies many different facets of interpretation. Using a single image to query a database might employ features that do not match user's expectation and retrieve results with low precision/recall ratios. How to automatically extract reliable image features as a query key that matches user's expectation in a content-based image retrieval (CBIR) system is an important topic.The objective of the present work is to propose a multiple-instance learning image retrieval system by incorporating an isometric embedded similarity measure. Multiple-instance learning is a way of modeling ambiguity in supervised learning given multiple examples. From a small collection of positive and negative example images, semantically relevant concepts can be derived automatically and employed to retrieve images from an image database. Each positive and negative example images are represented by a linear combination of fractal orthonormal basis vectors. The mapping coefficients of an image projected onto each orthonormal basis constitute a feature vector. The Euclidean-distance similarity measure is proved to remain consistent, i.e., isometric embedded, between any image pairs before and after the projection onto orthonormal axes. Not only similar images generate points close to each other in the feature space, but also dissimilar ones produce feature points far apart.The utilization of an isometric-embedded fractal-based technique to extract reliable image features, combined with a multiple-instance learning paradigm to derive relevant concepts, can produce desirable retrieval results that better match user's expectation. In order to demonstrate the feasibility of the proposed approach, two sets of test for querying an image database are performed, namely, the fractal-based feature extraction algorithm vs. three other feature extractors, and single-instance vs. multiple-instance learning. Both the retrieval results, execution time and precision/recall curves show favorably for the proposed multiple-instance fractal-based approach.     

A computation-oriented multimedia data streams model for content-based information retrieval

Multimedia applications nowadays are becoming prevalent. In the past the relational database model was generalized to the multimedia database model. More recently the relational database model was generalized to the data streams model, as the technology advanced and data became bulky and unbounded in size due to the utilization of sensor networks. In this paper we take one more step of generalization by providing a multimedia data streams model. The objective is to furnish a formal framework to design multimedia data streams (MMDS) schema for efficient content based information retrieval. We also extend the functional dependency theory and the normalization framework to handle multimedia data streams. Finally we present algorithmic methods of generating continuous multimedia queries along with examples for illustration.     

CORE: a content-based retrieval engine for multimedia information systems

Rapid advances in multimedia technology necessitate the development of a generic multimedia information system with a powerful retrieval engine for prototyping multimedia applications. We develop a content-based retrieval engine (CORE) that makes use of novel indexing techniques for multimedia object retrieval. We formalize the concepts related to multimedia information systems such as multimedia objects and content-based retrieval. We bring out the requirements and challenges of a multimedia information system. The architecture of CORE is described in detail along with the associated retrieval mechanisms and indexing techniques. Various modules developed for efficient retrieval are presented with some APIs. The efficacy of CORE is demonstrated in the development of two multimedia systems, a computer-aided facial image inference and retrieval (CAFIIR) system and a system for trademark archival and retrieval (STAR), which have been developed at the Institute of Systems Science (ISS). We expect that CORE will be useful for effective prototyping of other such multimedia applications.Mainly supported by National Science & Technology Board of SingaporePartly working in Real World Computing Partnership, Novel Function Institute of Systems Science Laboratory since April 1994.     

On stability of signature-based similarity measures for content-based image retrieval

Retrieving similar images from large image databases is a challenging task for today’s content-based retrieval systems. Aiming at high retrieval performance, these systems frequently capture the user’s notion of similarity through expressive image models and adaptive similarity measures. On the query side, image models can significantly differ in quality compared to those stored on the database side. Thus, similarity measures have to be robust against these individual quality changes in order to maintain high retrieval performance. In this paper, we investigate the robustness of the family of signature-based similarity measures in the context of content-based image retrieval. To this end, we introduce the generic concept of average precision stability, which measures the stability of a similarity measure with respect to changes in quality between the query and database side. In addition to the mathematical definition of average precision stability, we include a performance evaluation of the major signature-based similarity measures focusing on their stability with respect to querying image databases by examples of varying quality. Our performance evaluation on recent benchmark image databases reveals that the highest retrieval performance does not necessarily coincide with the highest stability.     

互信息启发的相似度组合图像检索算法

图像的视觉特征与用户描述之间的差距一直是影响基于内容的图像检索准确度的最主要因素。对多种相似度进行组合来检索图像是近几年图像检索领域涌现出的一个研究热点,也是缩小这种差距的一种有效途径。如何选择更好的组合方法则是该领域很多研究者关注的核心问题。提出一种新的相似度组合算法。该算法基于互信息度量相对熵的原理,计算连续变量相似度与离散变量相似性之间的相关性,对多种相似度进行选择,以“和规则”组合相似度。在公用数据集上进行检索实验,该算法优于当前其他的“和规则”下的组合方法。     

A middleware to enhance current multimedia retrieval systems with content-based functionalities

Nowadays the retrieval of multimedia assets is mainly performed by text-based retrieval systems with powerful and stable indexing mechanisms. Migration from those systems to content-aware multimedia retrieval systems is a common aim for companies from very diverse sectors. In this paper we present a semantic middleware designed to achieve a seamless integration with existing systems. This middleware outsources the semantic functionalities (e.g. knowledge extraction, semantic query expansion,…) that are not covered by traditional systems, thereby allowing the use of complementary content-based techniques. We include a list of key criteria to successfully deploy this middleware, which provides semantic support to many different steps of the retrieval process. Both the middleware and the design criteria are validated by two real complementary deployments in two very different industrial domains.     

A short-term learning approach based on similarity refinement in content-based image retrieval

Multimedia Tools and Applications - This paper presents a new relevance feedback approach based on similarity refinement. In the proposed approach weight correction of feature’s components is...     

MDCBIR-MF: multimedia data for content-based image retrieval by using multiple features

Multimedia Tools and Applications - Due to recent development in technology, the complexity of multimedia is significantly increased and the retrieval of similar multimedia content is an open...     

Learning non-metric visual similarity for image retrieval

Measuring visual similarity between two or more instances within a data distribution is a fundamental task in image retrieval. Theoretically, non-metric distances are able to generate a more complex and accurate similarity model than metric distances, provided that the non-linear data distribution is precisely captured by the system. In this work, we explore neural networks models for learning a non-metric similarity function for instance search. We argue that non-metric similarity functions based on neural networks can build a better model of human visual perception than standard metric distances. As our proposed similarity function is differentiable, we explore a real end-to-end trainable approach for image retrieval, i.e. we learn the weights from the input image pixels to the final similarity score. Experimental evaluation shows that non-metric similarity networks are able to learn visual similarities between images and improve performance on top of state-of-the-art image representations, boosting results in standard image retrieval datasets with respect standard metric distances.     

Association and content-based retrieval

In spite of important efforts in content-based indexing and retrieval during these last years, seeking relevant and accurate images remains a very difficult query. In the state-of-the-art approaches, the retrieval task may be efficient for some queries in which the semantic content of the query can be easily translated into visual features. For example, finding images of fires is simple because fires are characterized by specific colors (yellow and red). However, it is not efficient in other application fields in which the semantic content of the query is not easily translated into visual features. For example, finding images of birds during migrations is not easy because the system has to understand the query semantic. In the query, the basic visual features may be useful (a bird is characterized by a texture and a color), but they are not sufficient. What is missing is the generalization capability. Birds during migrations belong to the same repository of birds, so they share common associations among basic features (e.g., textures and colors) that the user cannot specify explicitly. We present an approach that discovers hidden associations among features during image indexing. These associations discriminate image repositories. The best associations are selected on the basis of measures of confidence. To reduce the combinatory explosion of associations, because images of the database contain very large numbers of colors and textures, we consider a visual dictionary that group together similar colors and textures.     

Localized content-based image retrieval

We define localized content-based image retrieval as a CBIR task where the user is only interested in a portion of the image, and the rest of the image is irrelevant. In this paper we present a localized CBIR system, Accio, that uses labeled images in conjunction with a multiple-instance learning algorithm to first identify the desired object and weight the features accordingly, and then to rank images in the database using a similarity measure that is based upon only the relevant portions of the image. A challenge for localized CBIR is how to represent the image to capture the content. We present and compare two novel image representations, which extend traditional segmentation-based and salient point-based techniques respectively, to capture content in a localized CBIR setting.     

Utilizing similarity relationships among existing data for high accuracy processing of content-based image retrieval

Retrieving similar images based on its visual content is an important yet difficult problem. We propose in this paper a new method to improve the accuracy of content-based image retrieval systems. Typically, given a query image, existing retrieval methods return a ranked list based on the similarity scores between the query and individual images in the database. Our method goes further by relying on an analysis of the underlying connections among individual images in the database to improve this list. Initially, we consider each image in the database as a query and use an existing baseline method to search for its likely similar images. Then, the database is modeled as a graph where images are nodes and connections among possibly similar images are edges. Next, we introduce an algorithm to split this graph into stronger subgraphs, based on our notion of graph’s strength, so that images in each subgraph are expected to be truly similar to each other. We create for each subgraph a structure called integrated image which contains the visual features of all images in the subgraph. At query time, we compute the similarity scores not only between the query and individual database images but also between the query and the integrated images. The final similarity score of a database image is computed based on both its individual score and the score of the integrated image that it belongs to. This leads effectively to a re-ranking of the retrieved images. We evaluate our method on a common image retrieval benchmark and demonstrate a significant improvement over the traditional bag-of-words retrieval model.     

A multi-level matching method with hybrid similarity for document retrieval

This paper presents a multi-level matching method for document retrieval (DR) using a hybrid document similarity. Documents are represented by multi-level structure including document level and paragraph level. This multi-level-structured representation is designed to model underlying semantics in a more flexible and accurate way that the conventional flat term histograms find it hard to cope with. The matching between documents is then transformed into an optimization problem with Earth Mover’s Distance (EMD). A hybrid similarity is used to synthesize the global and local semantics in documents to improve the retrieval accuracy. In this paper, we have performed extensive experimental study and verification. The results suggest that the proposed method works well for lengthy documents with evident spatial distributions of terms.