Clustering is the unsupervised classification of patterns (observations, data items, or feature vectors) into groups (clusters). The clustering problem has been addressed in many contexts and by researchers in many disciplines; this reflects its broad appeal and usefulness as one of the steps in exploratory data analysis. However, clustering is a difficult problem combinatorially, and differences in assumptions and contexts in different communities has made the transfer of useful generic concepts and methodologies slow to occur. This paper presents an overview of pattern clustering methods from a statistical pattern recognition perspective, with a goal of providing useful advice and references to fundamental concepts accessible to the broad community of clustering practitioners. We present a taxonomy of clustering techniques, and identify cross-cutting themes and recent advances. We also describe some important applications of clustering algorithms such as image segmentation, object recognition, and information retrieval.

/pdf/data-clustering-a-review-4c452f5cab.pdf

Data clustering: a review

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

From the Publisher: 
The accessible presentation of this book gives both a general view of the entire computer vision enterprise and also offers sufficient detail to be able to build useful applications. Users learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods. A CD-ROM with every copy of the text contains source code for programming practice, color images, and illustrative movies. Comprehensive and up-to-date, this book includes essential topics that either reflect practical significance or are of theoretical importance. Topics are discussed in substantial and increasing depth. Application surveys describe numerous important application areas such as image based rendering and digital libraries. Many important algorithms broken down and illustrated in pseudo code. Appropriate for use by engineers as a comprehensive reference to the computer vision enterprise.

Computer vision : a modern approach = 计算机视觉 : 一种现代的方法

Handwriting has continued to persist as a means of communication and recording information in day-to-day life even with the introduction of new technologies. Given its ubiquity in human transactions, machine recognition of handwriting has practical significance, as in reading handwritten notes in a PDA, in postal addresses on envelopes, in amounts in bank checks, in handwritten fields in forms, etc. This overview describes the nature of handwritten language, how it is transduced into electronic data, and the basic concepts behind written language recognition algorithms. Both the online case (which pertains to the availability of trajectory data during writing) and the off-line case (which pertains to scanned images) are considered. Algorithms for preprocessing, character and word recognition, and performance with practical systems are indicated. Other fields of application, like signature verification, writer authentification, handwriting learning tools are also considered.

/pdf/online-and-off-line-handwriting-recognition-a-comprehensive-4qmt9ty596.pdf

Online and off-line handwriting recognition: a comprehensive survey

IEEE Transactions on Pattern Analysis and Machine Intelligence

In this paper, a dynamic texture based compression scheme is devised for videos. Correspondence analysis is explored for the analysis of motion patterns in a video on the basis of optic flow data and then clusters of different motion patterns are created. Dynamic textures tend to disobey Horn and Schunck's assumption of brightness constancy and hence, optic flow residual is used as an indicator of their presence. The correspondence analysis results and optic flow residual are combined together in a new segmentation scheme. The optic flow data tracks the motion of groups of pixels to generate the flow lines. These flow lines are used in a synthesis scheme for creating an illusion of continuously flowing texture. The integration of this synthesis scheme in the compression format gives us considerable bit stream reduction corresponding to the dynamic texture regions. The scheme is integrated with standard H.264/AVC model, texture regions that do not fall under the above precinct of dynamic textures and non-texture regions are encoded-decoded by the H.264 model directly.

Dynamic texture synthesis for video compression

Promoting visual diversity in image search results has been of immense interest in the MIR (multimedia information retrieval) community. However, most of the work done in this space and the test collections used therein have ignored the use of social tags (user generated metadata) for image search result diversification on social multimedia platforms such as Flickr, Picasa etc. Unlike traditional multimedia content, content generated on such social media platforms are usually annotated with a rich set of explicit and implicit human generated metadata (referred here as social tags) such as keywords, textual description, category information, author's profile, user-to-user and user-to-content interaction etc which can be useful for the image search-result diversity task. In this paper we demonstrate how existing image search result diversification method can be extended to incorporate social tag information. Experiments on a real-world dataset shows that incorporating social tag features in some of the popular diversification algorithms results in improvement over baseline numbers.

Search result diversification in Flickr

We present a unique representation scheme for events in an area under surveillance, which provides a mechanism to analyze videos from multiple perspectives for unusual activity analysis. We propose clustering in event component spaces and define algebraic operations on these clusters to find co-occurrences of event components. A usualnessmeasure for clusters is proposed that not only gives a measure on how usual or unusual an activity is, but also a basis for analyzing and predicting the possibly usual or unusual activities that can occur in the surveillance region.

Unusual Activity Analysis in Video Sequences

In this work, we propose an image analogy based super-resolution technique that can be used as an effective tool for document image compression and multi-resolution viewing of the document. The technique uses Dugad and Ahuja method for resizing document images. Next, image analogies framework is applied to add the missing high frequency information. The encoder allows user to compress spatially lower resolution version of the image using any standard image compression technique, thus enabling substantial compression. At the decoder end, the image is resized using Dugad and Ahuja method and then enhanced using image analogy by appending the missing high frequency details, using a training pair of the same class of document image.

Image Analogy Based Document Image Compression

A connectionist network is presented for simultaneous perception of multiple categories. These categories provide an adequate explanation of the input features originating from multiple classes. The network optimises an appropriately defined error function for making the inference. A supervised learning algorithm is presented for learning the association between the features and each individual category. >

Santanu Chaudhury

Papers

Dynamic texture synthesis for video compression

Search result diversification in Flickr

Unusual Activity Analysis in Video Sequences

Image Analogy Based Document Image Compression

A connectionist network for simultaneous perception of multiple categories