Search

This paper proposes a new method for face recognition based on a multiresolution analysis tool called Digital Curvelet Transform. Multiresolution ideas notably the wavelet transform have been profusely employed for addressing the problem of face recognition. However, theoretical studies indicate, digital curvelet transform to be an even better method than wavelets. In this paper, the feature extraction has been done by taking the curvelet transforms of each of the original image and its quantized 4 bit and 2 bit representations. The curvelet coefficients thus obtained act as the feature set for classification. These three sets of coefficients from the three different versions of images are then used to train three Support Vector Machines. During testing, the results of the three SVMs are fused to determine the final classification. The experiments were carried out on three well known databases, viz., the Georgia Tech Face Database, AT&T "The Database of Faces" and the Essex Grimace Face Database. [ABSTRACT FROM AUTHOR]

The problem of 3D reconstruction of nonrigid objects from uncalibrated image sequences is addressed in the paper. Our method assumes that the nonrigid object is composed of rigid part and deformation part. We first recover the affine structure of the object and separate the rigid features from the deformation ones. Then stratify the structure from affine to Euclidean space by virtue of the rigid features. The novelty of the paper lies in two aspects. First, we propose a deformation weight constraint to the problem and prove that the recovered shape bases and shapes are transformation invariant under this constraint. Second, we propose a constrained power factorization (CPF) algorithm to recover the deformation structure in affine space. The algorithm overcomes the limitations of previous SVD-based methods and can work with missing data in the tracking matrix. Extensive experiments on synthetic data and real sequences validate the proposed method and show improvements over existing solutions. [ABSTRACT FROM AUTHOR]

Shape-based image retrieval is one of the most challenging aspects in Content-Based Image Retrieval (CBIR). A variety of techniques are reported in the literature that aim to retrieve objects based on their shapes; each of these techniques has its advantages and disadvantages. In this paper, we propose a novel scheme that exploits complementary benefits of several shape-based image retrieval techniques and integrates their assessments based on a pair-wise co-ranking process. The proposed scheme can handle any number of CBIR techniques; however, three common techniques are used in this study: Invariant Zernike Moments (IZM), Multi-Triangular Area Representation (MTAR), and Fourier Descriptor (FD). The performance of the proposed scheme is compared with that of each of the selected techniques. As will be demonstrated in this paper, the proposed co-ranking scheme exhibits superior performance. [ABSTRACT FROM AUTHOR]

In this paper, we approach the problem of determining the basic components from repetitive textured surfaces undergoing free-form deformations. Traditional methods for texture modeling are usually based on measurements performed on fronto-parallel planar surfaces. Recently, affine invariant descriptors have been proposed as an effective way to extract local information from non-planar texture surfaces. However, affine transformations are unable to model local image distortions caused by changes in surface curvature. Here, we propose a method for selecting the most representative candidates for the basic texture elements of a texture field while preserving the descriptors' affine invariance requirement. Our contribution in this paper is twofold. First, we investigate the distribution of extracted affine invariant descriptors on a nonlinear manifold embedding. Secondly, we describe a learning procedure that allows us to group repetitive texture elements while removing candidates presenting high levels of curvature-induced distortion. We demonstrate the effectiveness of our method on a set of images obtained from man-made texture surfaces undergoing a range of non-rigid deformations. [ABSTRACT FROM AUTHOR]

This paper presents a scheme for image decomposition and reconstruction, using complex Gabor wavelets. Gabor functions have been used extensively in areas related to the human visual system due to their localization in space and bandlimited properties. However, since the standard two-sided Gabor functions are not orthogonal and lead to nearly singular Gabor matrices, they have been used in the decomposition and feature extraction of images rather than in image reconstruction. In an attempt to reduce the singularity of the Gabor matrix and produce reliable image reconstruction, in this paper, we show that a single-sided Gabor function can accomplish both, with the reconstruction residual error being very small (PSNR of at least 300 dB). [ABSTRACT FROM AUTHOR]

Splicing systems are generative devices of formal languages, introduced by Head in 1987 to model biological phenomena on linear and circular DNA molecules. In this paper we introduce a special class of finite circular splicing systems named marked systems. We prove that a marked system S generates a regular circular language if and only if S satisfies a special (decidable) property. As a consequence, we show that we can decide whether a regular circular language is generated by a marked system and we characterize the structure of these regular circular languages. [ABSTRACT FROM AUTHOR]

This paper illustrates and extends an efficient framework, called the square-root-elastic (SRE) framework, for studying shapes of closed curves, that was first introduced in [2]. This framework combines the strengths of two important ideas - elastic shape metric and path-straightening methods - for finding geodesics in shape spaces of curves. The elastic metric allows for optimal matching of features between curves while path-straightening ensures that the algorithm results in geodesic paths. This paper extends this framework by removing two important shape preserving transformations: rotations and re-parameterizations, by forming quotient spaces and constructing geodesics on these quotient spaces. These ideas are demonstrated using experiments involving 2D and 3D curves. [ABSTRACT FROM AUTHOR]

The pupil localization algorithm is very important for a face recognition system. Traditional pupil localization algorithms are easy to be affected by uneven illuminations and accessories. Aiming at those limitations, a novel pupil localization algorithm is proposed in this paper. The algorithm firstly implements face image tilt adjustment and extracts eye region through the horizontal intensity gradient integral projection and Gabor filtering. Then in order to increase the eye detection accuracy, PCA is applied to select Gabor filter, and the projection enhancement algorithm is presented. At last the Negative Radial Symmetry is presented to locate the pupil position precisely in eye windows. Experimental results show that the method can locate the pupil position accurately, and demonstrate robustness to uneven lighting, noise, accessories and pose variations. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel approach to graph regularisation based on energy minimisation. Our method hinges in the use of a Ginzburg-Landau functional whose extremum is achieved efficiently by a gradient descend optimisation process. As a result of the treatment given in this paper to the regularisation problem, constraints can be enforced in a straightforward manner. This provides a means to solve a number of problems in computer vision and pattern recognition. To illustrate the general nature of our graph regularisation algorithm, we show results on two application vehicles, photometric stereo and image segmentation. Our experimental results demonstrate the efficacy of our method for both applications under study. [ABSTRACT FROM AUTHOR]

This paper presents a novel visual speech recognition approach based on motion segmentation and hidden Markov models (HMM). The proposed method identifies utterances from mouth video, without evaluating voice signals. The facial movements in the video data are represented using 2D spatial-temporal templates (STT). The proposed technique combines discrete stationary wavelet transform (SWT) and Zernike moments to extract rotation invariant features from the STTs. HMMs are used as speech classifier to model English phonemes. The preliminary results demonstrate that the proposed technique is suitable for phoneme classification with a high accuracy. [ABSTRACT FROM AUTHOR]

The lower the resolution of a given text is, the more difficult it becomes to segment and to recognize it. The resolution of screen-rendered text can be very low. With a typical x-height of 4 to 7 pixels it is much lower as in other low resolution OCR situations. Modern OCR approaches for such very low resolution text use a classification-based segmentation where the underlying classifier plays an important role. This paper presents a multiple classifier system for the classification of single characters. This system is used as a subsystem for the classification-based segmentation within a system to read screen-rendered text. The paper shows that the presented multiple classifier system outperforms the best former single classifier system on single characters by far and it shows the impact of using the multiple classifier system on the word reading performance. [ABSTRACT FROM AUTHOR]

This paper presents a performance evaluation of a set of fast block matching motion estimation methods of recent development. These methods are Kite-Cross-Diamond search (KCDS), Cross-Diamond-Hexagonal search (CDHS), Efficient-Hexagonal-Inner search (EHIS) and a newly proposed in this paper, called Cross-Half-Diamond low resolution search (CHDS), which is an improvement of Cross-Diamond-Hexagonal search. Also, Diamond search (DS) and Full search (FS) algorithms are implemented. These methods are evaluated based on the mean square error and the number of search points. Based on experimental results, it is concluded that DS, CDHS, CHDS and EHIS are appropriate for small and large motion contents, except EHIS for high frequencies and complex motion contents. The KCDS method has the lowest performance compared with other algorithms. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of estimating the 3D rigid pose of a CT volume of an object from its 2D X-ray projections. We use maximization of mutual information, an accurate similarity measure for multi-modal and mono-modal image registration tasks. However, it is known that the standard mutual information measure only takes intensity values into account without considering spatial information and its robustness is questionable. In this paper, instead of directly maximizing mutual information, we propose to use a variational approximation derived from the Kullback-Leibler bound. Spatial information is then incorporated into this variational approximation using a Gibbs random field model. The newly derived similarity measure has a least-squares form and can be effectively minimized by a multi-resolution Levenberg-Marquardt optimizer. Experimental results are presented on X-ray and CT datasets of a plastic phantom and a cadaveric spine segment. [ABSTRACT FROM AUTHOR]

The paper focusses on a group of segmentation problems dealing with 3D data sets showing thin objects that appear disconnected in the data due to partial volume effects or a large spacing between neighbouring slices. We propose a modification of the speed function for the well-known level set method to bridge these discontinuities. This allows for the segmentation of the object as a whole. In this paper we are concerned with treelike structures, particularly dendrites in microscopic data sets, whose shape is unknown prior to segmentation. Using the modified speed function, our algorithm segments dendrites and their spines, even if parts of the object appear to be disconnected due to artifacts. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new manifold representation capable of being applied for visual speech recognition. In this regard, the real time input video data is compressed using Principal Component Analysis (PCA) and the low-dimensional points calculated for each frame define the manifolds. Since the number of frames that from the video sequence is dependent on the word complexity, in order to use these manifolds for visual speech classification it is required to re-sample them into a fixed number of keypoints that are used as input for classification. In this paper two classification schemes, namely the k Nearest Neighbour (kNN) algorithm that is used in conjunction with the two-stage PCA and Hidden-Markov-Model (HMM) classifier are evaluated. The classification results for a group of English words indicate that the proposed approach is able to produce accurate classification results. [ABSTRACT FROM AUTHOR]

LSH (Locality Sensitive Hashing) is one of the best known methods for solving the c-approximate nearest neighbor problem in high dimensional spaces. This paper presents a variant of the LSH algorithm, focusing on the special case of where all points in the dataset lie on the surface of the unit hypersphere in a d-dimensional Euclidean space. The LSH scheme is based on a family of hash functions that preserves locality of points. This paper points out that when all points are constrained to lie on the surface of the unit hypersphere, there exist hash functions that partition the space more efficiently than the previously proposed methods. The design of these hash functions uses randomly rotated regular polytopes and it partitions the surface of the unit hypersphere like a Voronoi diagram. Our new scheme improves the exponent ρ, the main indicator of the performance of the LSH algorithm. [ABSTRACT FROM AUTHOR]

In the last decade there has been an ongoing interest in string comparison problems; to a large extend the interest was stimulated by genome rearrangement problems in computational biology but related problems appear in many other areas of computer science. Particular attention has been given to the problem of sorting by reversals(SBR): given two strings, A and B, find the minimum number of reversals that transform the string A into the string B (a reversalρ(i,j), i

This paper introduces a new benchmark study of evaluating landmark-based shape correspondence used for statistical shape analysis. Different from previous shape-correspondence evaluation methods, the proposed benchmark first generates a large set of synthetic shape instances by randomly sampling a specified ground-truth statistical shape model. We then run the test shape-correspondence algorithms on these synthetic shape instances to construct a new statistical shape model. We finally introduce a new measure to describe the difference between this newly constructed statistical shape model and the ground truth. This new measure is then used to evaluate the performance of the test shape-correspondence algorithm. By introducing the ground-truth statistical shape model, we believe the proposed benchmark allows for a more objective evaluation of the shape correspondence than those that do not specify any ground truth. [ABSTRACT FROM AUTHOR]

In this paper, the performance of a wireless communication system based on the TD-CDMA transmission technique is analyzed. In this paper, we present simulation results for the performance of a turbo coded TD-CDMA system with QPSK over Rayleigh fading channel model. And the system performance employing the simple averaging channel estimation is compared to the ideal case with the perfect channel estimation. [ABSTRACT FROM AUTHOR]

Feature extraction is one of the most important problems in face recognition task. In this paper, we use kernel uncorrelated discriminant analysis to extract the optimal discriminant features for face recognition. The method also solves the so-called "Small Sample Size" (SSS) problem, which exists in most Face Recognition tasks. Experimental results on the Yale face database and AT&T face database show the effectiveness of this method. [ABSTRACT FROM AUTHOR]

In this study, the surveyed Trip Length Frequency Distribution (TLFD)is determined as a criterion for the reliability of evaluating the true O/D matrix. The surveyed TLFD can be used to check the similarity between the surveyed (true) Trip Length Distribution and the Trip Length Distribution of the estimated O/D matrix by the traffic counted models. When the surveyed TLFD is similar to the estimated TLFD, the reliability and correctness of the estimated O/D are high. Therefore, the objective of this paper is the development of the travel demand (O/D matrix) estimation using traffic counts on the large-scaled network. The Gradient Method is used for the model and the multi-class assignment technique is used for the equilibrium loading procedure in the model. This leads to the good guideline to the usage of the traffic count based O/D estimation in practice and gives a confidence to the transport planner. It is because the traffic counted O/D estimation models gives multiple solutions by its characteristics. In this paper we analyze the merits and demerits in each of a single-class based model and a multi-class based model in a large scale network. As a result, we have concluded that the multi-class based model has a closer value to the surveyed (true) TLFD than the TLFD of the estimated O/D matrix by the single-class based gradient method. [ABSTRACT FROM AUTHOR]

This paper develop the complexity reduction schemes of real root method that is mainly used in the CELP (Code Excited Linear Prediction) vocoder. The real root method is that if polynomial equations have the real roots, it is able to find those and transform them into LSP (Line Spectrum Pairs). Proposed algorithm is developed by using Mel scale and is to reduce the LSP complexity That is, the searching interval is arranged by using Mel scale but not it is uniform. In experimental results, complexity of the developed algorithm is reduced about 46% in average, but the transformed LSP parameters of the proposed method were the same as those of real root method. Hence, in case of applying proposed algorithm in G.723.1 (6.3kbps MP-MLQ), the speech quality is no distortion compared to original speech quality. [ABSTRACT FROM AUTHOR]

With the rapid development of wireless multimedia communication and Peer-to-Peer overlay network, it is envisioned that building a content and location addressable overlay network for wireless multimedia communication is promising. In this paper, based on a special structure, Geographically Hierarchical Index (GHI), we propose an efficient algorithm, Content and Location Addressable overlay Network (CLAN) algorithm, which makes p2p multimedia communication over MANETs applicable and efficient, satisfying three goals: efficiency, scalability and adaptability to node movement. [ABSTRACT FROM AUTHOR]

A music structured query language, called MuSQL, is presented in this paper. MuSQL consists of a schema definition sub-language and a data manipulation sub-language. The former is composed of schema-setup statements, schema-alter statements, and schema-drop statements. The latter is composed of selection, retrieval, extraction, insertion, update, deletion, commission, rollback, and other statements. MuSQL can be used to cut, delete and merge content of music, insert, delete and extract features of music, and exactly or approximately search music pieces, especially in the processing of music based on content. Also, it makes some music processing operations easier due to its built-in semantics. MuSQL has been implemented in a music data management system. [ABSTRACT FROM AUTHOR]

In this paper, we propose fast color interpolation methods for color filter array (CFA). The proposed methods exploit correlations among the three color components. It has been reported that the quality of color interpolation can be significantly improved by utilizing correlation among color components. However, the complexity of such methods may increase significantly compared to those methods which separately interpolate each color component. On the other hand, the major application of CFA is portable imaging devices such as digital cameras or mobile phones. For portable devices, power consumption is an important issue. In this paper, we propose high quality color interpolation methods with reduced complexity. Experimental results show that the proposed methods provide noticeable improvements without significantly increasing the system complexity. [ABSTRACT FROM AUTHOR]

3D facial synthesis has been an intensive research topic in both image processing and computer graphics. So far common facial synthesizing methods were either statistic model based or laser range scanner (LRS) based. However, these methods could only provide approximate models. In this paper, we present a Magnetic Resonance Image (MRI) based 3D facial synthesizing approach, which can build accurate virtual faces for plastic surgery. This approach provides an database free modeling process so that the precision will not be limited by the face database. During modeling, we combine Sobel filter with snake algorithm to extract the 3D facial surface from noised MRI and build Bezier facial model, which facilitates the local geometric control. Furthermore, the proposed photo mapping algorithm help add realistic texture to the model. It avoids the complex and time-consuming texture fitting process while achieves satisfying result. [ABSTRACT FROM AUTHOR]

This paper proposes a robust point transferring method for markerless AR applications. Using this method, any points specified at the initialization stage can be stably transferred during the augmentation process. These transferred points can be used for registration, annotation and video augmentation in markerless AR applications. This proposed point transferring method is based on a simple nonlinear optimization model. The proposed method has several advantages. Firstly, it is robust and stable as it remains effective when the camera is moved about quickly or when the scenes are largely occluded or filled with moving objects. Second, it is simple as the points that will be used for registration, annotation and video augmentation are only required to be specified in one image. Lastly, it is fast as the proposed simple optimization model can be solved quickly. Several experiments have been conducted to validate the performance of this proposed method. [ABSTRACT FROM AUTHOR]

The purpose of an image hash is to provide a compact representation of the whole image. Designing a good image hash function requires careful consideration of many issues such as robustness, security and tamper detection with precise localization. In this paper, we present a novel hashing scheme that addresses these issues in a unified framework. We analyze the security issues in image hashing and present new ideas to counter some of the attacks that we shall describe in this paper. Our proposed scheme is resilient to allow non-malicious manipulations like JPEG compression, high pass filtering and is sensitive enough to detect tampering with precise localization. Several experimental results are presented to demonstrate the effectiveness of the proposed scheme. [ABSTRACT FROM AUTHOR]

Energy consumption is a key issue in modern microprocessor system design in general, and in the design of mobile computing devices more in particular. This paper introduces a novel approach to energy-efficient media stream decoding that is based on the notion of media stream similarity. The key idea is that platform-independent scenarios of similar decode complexity can be identified within and across media streams. A client decoding a media stream annotated with scenario information can then adjust its processor clock frequency and voltage level based on these scenarios for reduced energy consumption. Our evaluation done using the AVC decoder and 12 reference streams shows an average energy reduction of 46% while missing less than 0.2% of the frame deadlines on average. [ABSTRACT FROM AUTHOR]

Background Subtraction methods are wildly used to detect moving object from static cameras. It has many applications such as traffic monitoring, human motion capture and recognition, and video surveillance. It is hard to propose a background model which works well under all different situations. Actually, there is no need to propose a pervasive model; it is a good model as long as it works well under a special situation. In this paper, a new method combining Gaussian Average and Frame Difference is proposed. Shadow suppression is not specifically dealt with, because it is considered to be part of the background, and can be subtracted by using an appropriate threshold. At last, a new method is raised to fill small gaps that the detected foreground or the moving objects may contain. [ABSTRACT FROM AUTHOR]

This paper presents a neuro-fuzzy classifer for activity recognition using one triaxial accelerometer and feature reduction approaches. We use a triaxial accelerometer to acquire subjects' acceleration data and train the neuro-fuzzy classifier to distinguish different activities/movements. To construct the neuro-fuzzy classifier, a modified mapping-constrained agglomerative clustering algorithm is devised to reveal a compact data configuration from the acceleration data. In addition, we investigate two different feature reduction methods, a feature subset selection and linear discriminate analysis. These two methods are used to determine the significant feature subsets and retain the characteristics of the data distribution in the feature space for training the neuro-fuzzy classifier. Experimental results have successfully validated the effectiveness of the proposed classifier. [ABSTRACT FROM AUTHOR]

Films have been the most entertaining art form during the past century. Sometimes they were inspired in written novels; sometimes they have inspired new written novels. Film scripts are halfway between the film in the screen and the pure world of written imagination. Real time is one of the dimensions lost in the script, breaking the anchors to the time signals of what films are made. This paper presents a full approach for merging these two worlds in the real time dimension. Using subtitles time stamping and a new parallel text alignment algorithm, a time stamped script is produced. This is used to create new enriched narrative films, also presented in the paper. [ABSTRACT FROM AUTHOR]

We present an interactive image-based technique to relight real scene with physical light sources acquisition. Firstly, basis images of the real scene are acquired from a fixed viewpoint. Secondly, light direction is estimated and physical light image is captured. Thirdly, measurement image between the reference light image and the novel light image is computed for relighting scene. We demonstrate the technique by interactively relighting a pre-captured real scene using a hand held spotlight projecting light on a white paper. [ABSTRACT FROM AUTHOR]

One of the most exciting aspects of shape modeling is the development of new algorithms and methods to create unusual, interesting and aesthetically pleasing shapes. In this paper, we present an interactive modeling system for generating freeform surfaces using a 2D user interface. In this paper, we firstly interpret the given 2D curve to be the projection of the 3D curve that has the minimum curvature. Then we propose a topology-consistent strategy to trace the simple faces on the interconnecting 3D curves. By using the face tracing algorithm, our system can identify the 3D surfaces automatically. After obtaining the boundary curves for the faces, we apply Delaunay triangulation on these faces. Finally, the shape of the triangule surface mesh that follows the 3D boundary curves is computed by using harmonic interpolation. Solving Laplace's equations, we are able to construct simple and conceptual 3D models interactively. Although the incorporation of topological manipulation will introduce many new(and interesting) design problems that cannot be addressed in this paper, we show that automatically generated models are both beneficial and feasible. [ABSTRACT FROM AUTHOR]

Highly detailed models are becoming commonplace in computer graphics. Such models are often represented as triangle meshes, challenging rendering performance, transmission bandwidth, and storage capacities. This paper proposes a new scheme for constructing progressive meshes with adaptive subdivision. Besides, we introduce half-edge collapse simplification and present a new way to calculate error distance. Some experimental results show that our proposed method works well on various meshes. [ABSTRACT FROM AUTHOR]

In the past, many tabletop systems have used the traditional table augmented with digital technologies to enhance collaborative works such as face-to-face or remote meeting and working. In this paper, we describe the Tea Table Mediator, a multimodal ambient tabletop display designed for five senses convergence experience combined with tea drinking. This ambient display enhances a traditional tea table with multimodal interfaces forintimate small group interaction. This paper describes the design considerations, the details on system implementation, and the discussion of a preliminary user evaluation. [ABSTRACT FROM AUTHOR]

Multidimensional databases are commonly used for decision making in the context of data warehouses. Considering the multidimensional model, data are presented as hypercubes organized according to several dimensions. However, in general, hypercubes have more than three dimensions and contain a huge amount of data, and so cannot be easily visualized. In this paper, we show that data cubes can be visualized as images by building blocks that contain mostly the same value. Blocks are built up using an APriori-like algorithm and each block is considered as a set of pixels which colors depend on the corresponding value. The key point of our approach is to set how to display a given block according to its corresponding value while taking into account that blocks may overlap. In this paper, we address this issue based on the Pixelization paradigm. [ABSTRACT FROM AUTHOR]

Head pose estimation in low resolution is a challenge problem. Traditional pose estimation algorithms, which assume faces have been well aligned before pose estimation, would face much difficulty in this situation, since face alignment itself does not work well in this low resolution scenario. In this paper, we propose to estimate head pose using view-based multi-view face detectors directly. Naive Bayesian classifier is then applied to fuse the information of head pose from multiple camera views. To model the temporal changing of head pose, Hidden Markov Model is used to obtain the optimal sequence of head pose with greatest likelihood. [ABSTRACT FROM AUTHOR]

In this paper, we present KNIGHT, a Windows-based stand-alone object detection, tracking and classification software, which is built upon Microsoft Windows technologies. The object detection component assumes stationary background settings and models background pixel values using Mixture of Gaussians. Gradient-based background subtraction is used to handle scenarios of sudden illumination change. Connected- component algorithm is applied to detected foreground pixels for finding object-level moving blobs. The foreground objects are further tracked based on a pixel-voting technique with the occlusion and entry/exit reasonings. Motion correspondences are established using the color, size, spatial and motion information of objects. We have proposed a texture-based descriptor to classify moving objects into two groups: vehicles and persons. In this component, feature descriptors are computed from image patches, which are partitioned by concentric squares. SVM is used to build the object classifier. The system has been used in the VACE-CLEAR evaluation forum for the vehicle tracking task. Corresponding system performance is presented in this paper. [ABSTRACT FROM AUTHOR]

The H.264 video coding standard provides considerably higher coding efficiency than those of previous standards but its complexity is significantly increased. In this paper, we propose an efficient method of fast mode decision by exploiting spatio-temporal correlation in H.264. Firstly, we select skip mode or inter mode by considering the temporal correlation. Secondly, we select variable block size on inter mode by considering the spatial correlation. Simulations show that the proposed method reduces the encoding time by 71% on average without any significant PSNR losses. [ABSTRACT FROM AUTHOR]

As the amount of visual information is rapidly increasing, users want to find the more semantic information easily. Most retrieval systems by low-level features(such as color, texture) could not satisfy user's demand. To interpret semantic of image, many researchers use keywords as textual annotation. However, it's the image retrieval without ranking by text matching which is the simplest way to retrieval according to keyword's existence or nonexistence. In this paper, we propose conceptualization by similarity measure using relations among keywords for efficient image retrieval. We experiment annotated image retrieval by lowering the unrelated keyword's weight value and raising important keyword's one. [ABSTRACT FROM AUTHOR]

Traditional boosting method like adaboost, boosts a weak learning algorithm by updating the sample weights (the relative importance of the training samples) iteratively. In this paper, we propose to integrate feature re-weighting into boosting scheme, which not only weights the samples but also weights the feature elements iteratively. To avoid overfitting problem caused by feature re-weighting on a small training data set, we also incorporate relevance feedback into boosting and propose an interactive boosting called i.Boosting. It merges adaboost, feature re-weighting and relevance feedback into one framework and exploits the favorable attributes of these methods. In this paper, i.Boosting is implemented using Adaptive Discriminant Analysis (ADA) as base classifiers. It not only enhances but also combines a set of ADA classifiers into a more powerful one. A feature re-weighting method for ADA is also proposed and integrated in i.Boosting. Extensive experiments on UCI benchmark data sets, three facial image data sets and COREL color image data sets show the superior performance of i.Boosting over AdaBoost and other state-of-the-art projection-based classifiers. [ABSTRACT FROM AUTHOR]

Scalar multiplication dP and gP+hQ are important in encryption, decryption and signature in information security and wireless network. The speed of computation of scalar multiplication is significant for related applications. In this paper, a new signed binary representation (SBR) for integers called complementary code method (CC) is proposed, which has minimum weight and needs less memory. An efficient algorithm using CC method for computing dP is shown also. According to analyzing and comparing to the other methods, this algorithm is the better one in window methods and is the simplest for applying in software and hardware. By applying joint representation in computing gP+hQ, new algorithm using CC method has the least joint weight compared to other methods mentioned in this paper. So, the new SBR can efficiently speed up the computation of scalar multiplication dP and gP+hQ and can be widely used in secure communication for improving the speed of encryption and signature. [ABSTRACT FROM AUTHOR]

A novel embedded processor element basing on the Transport Triggered Architecture is presented in this paper. The processor element consisting of two powerful arithmetic clusters using the application specific instruction processor design methodology achieves higher performance and is especially good at exploiting the instruction level and data level parallelisms in the multimedia applications. To improve the efficiency, the processor also presents the decoupled stream memory system with the characteristics of the stream buffer proxy to support the cross-line indexed accesses and to enhance the memory bandwidth. Then, a heterogeneous multiprocessor SoC chip involving the embedded processor is fabricated using 0.13um CMOS process, and the SoC operates at 400MHz and consumes only around 690mW. Experimental results show that the embedded processor element has good performance improvement for the multimedia applications. [ABSTRACT FROM AUTHOR]

In recent years, with the rapid growth of the Internet as well as the increasing demand for broadband services, live pay-television broadcasting via internet has become a promising business. To get this implemented, it is necessary to protect distributed contents from illegal copying and redistributing after they are accessed. Fingerprinting system is a useful tool for it. This paper shows that the anti-collusion code has advantages over other existing fingerprinting codes in terms of efficiency and effectivity for live pay-television broadcasting. Next, this paper presents how to achieve efficient and effective anti-collusion codes based on affine plane and unital, which are two known examples of balanced incomplete block design (BIBD). Meanwhile, performance evaluations of anti-collusion codes generated from affine plane and unital are conducted. Their practical explicit constructions are given last. [ABSTRACT FROM AUTHOR]

This paper describes a novel object mining system for videos. An algorithm published in a previous paper by the authors is used to segment the video into shots and extract stable tracks from them. A grouping technique is introduced to combine these stable tracks into meaningful object clusters. These clusters are used in mining similar objects. Compared to other object mining systems, our approach mines more instances of similar objects in different shots. The proposed framework is applied to a full length feature film and improved results are shown. [ABSTRACT FROM AUTHOR]

By reviewing DRMs up to now and two typical examples - AVS DRM and DMP IDP, the paper tries to find out the fundamental challenge of content protection approach from technical and social viewpoints. Not only it is difficult to deploy and update content encryption and security infrastructure, but also the content diffusion is limited and Fair Use is affected. The new schema for DRM should be content monitoring system in public space that prevents illegal diffusion of content in copyright but permits content being used freely in private space or for social liberty. The traditional rights in analog times will fluently move to digital space under the proposed schema. [ABSTRACT FROM AUTHOR]

In this presentation for the panel at MCAM07, I put forward the transition of modeling the world as was done on a large scale in computer vision before the year 2000, to the current situation where there have been considerable successes with multimedia analysis by learning from the world. We make a plead for the last type of learned features, modeling only the scene accidental conditions and learning the object or object class intrinsic properties. In this paper, in respect to contributions by many others, we illustrate the approach of learning features by papers from our lab at the University of Amsterdam. [ABSTRACT FROM AUTHOR]

Although content-based image retrieval has been studied for decades, most commercial image search engines are still text-based. However, there is a growing demand for techniques to support content-based image search at Web scale. In this paper, we propose an ambitious goal to searching one billion Web images by content, and discuss the major challenges and opportunities. We also present several important applications that can be greatly benefited by techniques to enable Web-scale image search by content. These applications include image copyright infringement detection, street-side photo search, and search-based image annotation. We believe that the insights presented in the paper are enlightening to researchers in this field, and any breakthrough we make in this space will lead to many impactful applications in the future. [ABSTRACT FROM AUTHOR]

Tree-based algorithms, such as Patricia, LC-trie, LPFST, etc, are widely used to do longest prefix matching (LPM). These algorithms use all the bits of the prefix to build the tree and the bits are used from the most significant bit to the least significant bit sequentially. Thus the tree is not balanced and the tree depth is high. In this paper, we propose bit selection tree (BS-tree) to do LPM. The bits of the prefix are selected to build BS-tree based on their costs defined in this paper. BS-tree has lower tree depth and is more balanced compared to other tree-based schemes. BS-tree has good scalability to the length of the IP address and is suitable for both IPv4 and IPv6. We evaluate the performances of BS-tree using both IPv4 and IPv6, and specially refine it for IPv6 based on the observations on IPv6 address and real IPv6 routing tables. [ABSTRACT FROM AUTHOR]