As person parts are frequently misaligned between detected human boxes, an image representation that can handle this part misalignment is required. In this work, we propose an effective grouped attentive re-identification (GAReID) framework to learn part-aligned and background robust representations for person re-identification (ReID). Specifically, the GAReID framework consists of grouped high-order pooling (GHOP) and attentive high-order pooling (AHOP) layers, which generate high-order image and foreground features, respectively. In addition, a novel grouped Kronecker product (GKP) is proposed to use both channel group and shuffle strategies for high-order feature compression, while promoting the representational capabilities of compressed high-order features. We show that our method derives from an interpretable motivation and elegantly reduces part misalignments without using landmark detection or feature partition. This article theoretically and experimentally demonstrates the superiority of the GAReID framework, achieving state-of-the-art performance on various person ReID datasets.

In this paper we propose a framework using multi-channel convolutional neural network (MC–CNN) for recognizing the grammatical class (verb or noun) of covertly-spoken words from electroencephalogram (EEG) signals. Our proposed network extracts features by taking into account spatial, temporal, and spectral properties of the EEG signal. Further, sets of signals acquired from different regions of the brain are processed separately within the proposed framework and are subsequently combined at the classification stage. This approach enables the network to effectively learn discriminative features from the locations of the brain where imagined speech is processed. Our network was tested using challenging experiments, including cases where the test subject did not take part in system training. In our main application scenario, where no instance of a specific noun or verb was used during training, our method achieved 85.7% recognition. Further, our proposed method was evaluated on a publicly available EEG dataset and achieved recognition rate of 93.8% in binary classification. These results demonstrate the potential of our method.

Despite the remarkable progress in recent years, person Re-Identification (ReID) approaches frequently fail in cases where the semantic body parts are misaligned between the detected human boxes. To mitigate such cases, we propose a novel High-Order ReID (HOReID) framework that enables semantic pose alignment by aggregating the fine-grained part details of multilevel feature maps. The HOReID adopts a high-order mapping of multilevel feature similarities in order to emphasize the differences of the similarities between aligned and misaligned part pairs in two person images. Since the similarities of misaligned part pairs are reduced, the HOReID enhances pose-robustness within the learned features. We show that our method derives from an intuitive and interpretable motivation and elegantly reduces the misalignment problem without using any prior knowledge from human pose annotations or pose estimation networks. This paper theoretically and experimentally demonstrates the effectiveness of the proposed HOReID, achieving superior performance over the state-of-the-art methods on the four large-scale person ReID datasets.

We explore the capabilities of a new biometric trait, which is based on information extracted through facial motion amplification. Unlike traditional facial biometric traits, the new biometric does not require the visibility of facial features, such as the eyes or nose, that are critical in common facial biometric algorithms. In this paper we propose the formation of a spatiotemporal facial blood flow map, constructed using small motion amplification. Experiments show that the proposed approach provides significant discriminatory capacity over different training and testing days and can be potentially used in situations where traditional facial biometrics may not be applicable.

Autostereoscopic 3D display technology provides a perception of natural 3D effects without requiring viewers to wear any headgear device. In particular, Holoscopic 3D display mimics fly's eye using a microlens array in order to reconstruct a real scene in the 3D space and offers a fatigue-free viewing experience. To date, holoscopic 3D displays have been developed based on LCD technology. In this paper, we propose a LED holoscopic 3D wall-size display with effective pixel mapping and microlens parameters which deliver a long-distance 3D viewing experience. Due to the growing popularity of large LED displays, the proposed technology has significant potential for deployment in the advertisement and entertainment sectors.

We propose a new biometric trait based on facial motion amplification. The main advantage of the new biometric characteristic is that it does not rely on the visibility of critical facial features, such as nose, mouth, iris, or eyebrows. This makes it effective even when the respective areas are covered. Using the proposed system, facial image sequences are captured using an ordinary video camera and facial blood flow is calculated by means of small motion amplification. The calculated blood flow is captured from limited facial areas and is represented as a template that is suitable for identification purposes. Experiments on a new database show promising performance of the proposed approach, and provide evidence of the discriminatory capacity of the proposed biometric.

In this paper we introduce a novel representation for the classification of 3D images. Unlike most current approaches, our representation is not based on a fixed pyramid but adapts to image content and uses image regions instead of rectangular pyramid scales. Image characteristics, such as depth and color, are used for defining regions within images. Multiple region scales are formed in order to construct the proposed pyramid image representation. The proposed method achieves excellent results in comparison to conventional representations.

Presents a listing of the new Associate Editors for this issue of the publication.

We introduce a novel classification method for pyramid image representations that is particularly efficient when a small training set is available. A pyramid image representation is usually a unique concatenation of pyramid scale vectors of different discriminatory power. We propose training and classification on a scale by scale basis, i.e., using multiple vectors for the representation of an image. The proposed approach achieves approximately equal performance in comparison to the conventional approach but requires much smaller training sets. It also achieves excellent results for three-dimensional image classification.

This paper proposes a semi-supervised nonnegative matrix factorization algorithm for face and gait recognition. The proposed algorithm imposes hard constraints on the labelled data points, such that the data points that belong to the same class are projected to the same lower dimensional point. In addition, it introduces a graph Laplacian regularization term that preserves the local geometry structure of the data by penalising large distances between the projections of points that are close in the original space. This results in a constrained optimization problem, that is solved using block coordinate descent with multiplicative update rules. Experimental results on several publicly available datasets demonstrate that proposed method performs in par or considerably better than state of the art methods.

We propose a new gait recognition method that combines holistic and model-based features. Both types of features are extracted automatically from gait silhouette sequences and their combination takes place by means of a pair of hidden Markov models. In the proposed system, the holistic features are initially used for capturing general gait dynamics whereas, subsequently, the model-based features are deployed for capturing more detailed sub-dynamics by refining upon the preceding general dynamics. Furthermore, the holistic and model-based features are suitably processed in order to improve the discriminatory capacity of the final system. The experimental results show that the proposed method exhibits performance advantages in comparison with popular existing methods.

In a previous paper, Chang et al. presented a method for iterative decoding of FEC-based multiple description codes in image transmission. In this correspondence, we clarify that an outer interleaver used in the above research was previously proposed for the iterative decoding and optimization of product codes in image transmission.

In this paper, we present a novel and efficient gait recognition system. The proposed system uses two novel gait representations, i.e., the shifted energy image and the gait structural profile, which have increased robustness to some classes of structural variations. Furthermore, we introduce a novel method for the simulation of walking conditions and the generation of artificial subjects that are used for the application of linear discriminant analysis. In the decision stage, the two representations are fused. Thorough experimental evaluation, conducted using one traditional and two new databases, demonstrates the advantages of the proposed system in comparison with current state-of-the-art systems.

In this paper, we present a gait recognition method that does not presume the existence of strict lab conditions for its operation. The proposed algorithm includes a side-view detection and extraction approach that is useful when the subject is walking randomly as well as a novel template for gait representation that is robust to body posture variations. Experimental results show that the proposed system not only has high computational efficiency, but also exhibits robust performance, especially in cases where the subject is walking along random paths.

In this paper, we present a novel method for the simulation of walking conditions and the generation of artificial subjects that are used for the application of Linear Discriminant Analysis. The proposed method works efficiently in situations where only one gallery sequence with one gait cycle is available for each subject in the database. The proposed method was experimentally assessed in combination with the Gait Energy Image (GEI) and the Shifted Energy Image (SEI). Very considerable improvements on the recognition and verification results are achieved.

In this paper, we investigate the fusion of several features extracted from manually-labelled silhouettes. A novel approach for human gait recognition based on the combination of three discriminative features, i.e., the area, the gravity centre, and the orientation of each body component, is also proposed. Experimental results show that the proposed method exhibits considerably better performance, in comparison to all existing methods that use manually-extracted silhouettes.

This paper proposes a gait recognition system in which several views are available. It is shown that each view has unequal discrimination power and, therefore, should have unequal contribution in the recognition process. In order to exploit the availability of multiple views, several methods for the combination of the results that are obtained from the individual views are tested and evaluated. A novel approach for the combination of the results from several views is also proposed based on the relative importance of each view. The proposed approach generates improved results, compared to those obtained by using individual views or by using multiple views that are combined using other combination methods.

This paper presents a novel approach for gait recognition based on the matching of body components. The human body components are studied separately and are shown to have unequal discrimination power. Several approaches are presented for the combination of the results obtained from different body components into a common distance metric for the evaluation of similarity between gait sequences. A method is also proposed for the determination of the weighting of the various body components based on their contribution to recognition performance. Using the best performing of the proposed methods, improved recognition performance is achieved.

This paper proposes a robust object segmentation algorithm that tackles problems that arise in environments in which the foreground and background colours are similar and there is light reflection in the shadow areas. The proposed algorithm bases its efficiency on a novel RGB colour detection process with adaptive threshold and edge detection, which are combined in order to obtain a foreground map containing objects and shadows. Subsequently, edge information is used to remove the shadow regions that are out of the edge bounding boxes. Finally, a post-processing procedure is applied and, for offline detection, a temporal filter is included in order to retrieve the misclassified pixels. The experimental results demonstrate the superior performance of our algorithm in comparison to other existing methods.

This paper addresses the problem of frame interpolation for Wyner-Ziv video coding. A novel frame interpolation method based on block-adaptive matching algorithm for motion estimation is presented. This scheme enables block size adaptation to local activity within frames using block merging and splitting techniques. The efficiency of the proposed method is evaluated in transform domain Wyner-Ziv video coding. The experimental results demonstrate the superiority of the proposed method over existing frame interpolation techniques.

A new feature extraction process is proposed for gait representation and recognition. The new system is based on the Radon transform of binary silhouettes. For each gait sequence, the transformed silhouettes are used for the computation of a template. The set of all templates is subsequently subjected to linear discriminant analysis and subspace projection. In this manner, each gait sequence is described using a low-dimensional feature vector consisting of selected Radon template coefficients. Given a test feature vector, gait recognition and verification is achieved by appropriately comparing it to feature vectors in a reference gait database. By using the new system on the Gait Challenge database, very considerable improvements in recognition performance are seen in comparison to state-of-the-art methods for gait recognition.

A new process is proposed for gait representation and recognition. The new system is based on the Radon transform of binary silhouettes. For each gait sequence, the transformed silhouettes are used for the computation of a template. The set of all templates is subsequently subjected to linear discriminant analysis and subspace projection. In this manner, each gait sequence is described using a low-dimensional feature vector consisting of selected Radon template coefficients. By using the new system on the gait challenge database, very considerable improvements in recognition performance are seen in comparison to state-of-the-art methods for gait recognition.

We propose a novel scheme for error resilient image transmission. The proposed scheme employs a product coder consisting of LDPC codes and RS codes in order to deal effectively with bit errors. The efficiency of the proposed scheme is based on the exploitation of determinate symbols in Tanner graph decoding of LDPC codes and a novel product code optimization technique based on error estimation. Experimental evaluation demonstrates the superiority of the proposed system in comparison to recent state-of-the art techniques for image transmission.

We present a novel scheme for the transmission of H.264/AVC video streams over lossy packet networks. The proposed scheme exploits the error resilient features of H.264/AVC codec and employs Reed-Solomon codes to protect effectively the streams. The optimal classification of macroblocks into slice groups and the optimal channel rate allocation are achieved by iterating two interdependent steps. Simulations clearly demonstrate the superiority of the proposed method over other recent algorithms for transmission of H.264/AVC streams

The transmission of JPEG2000 images over wireless channels is examined using reorganization of the compressed images into error-resilient, product-coded streams. The product-code consists of Turbo-codes and Reed-Solomon codes which are optimized using an iterative process. The generation of the stream to be transmitted is performed directly using compressed JPEG2000 streams. The resulting scheme is tested for the transmission of compressed JPEG2000 images over wireless channels and is shown to outperform other algorithms which were recently proposed for the wireless transmission of images.

We propose a novel scheme for error-resilient image transmission. The proposed scheme employs a product coder consisting of low-density parity check (LDPC) codes and Reed-Solomon codes in order to deal effectively with bit errors. The efficiency of the proposed scheme is based on the exploitation of determinate symbols in Tanner graph decoding of LDPC codes and a novel product code optimization technique based on error estimation. Experimental evaluation demonstrates the superiority of the proposed system in comparison to recent state-of-the-art techniques for image transmission.

A novel image transmission scheme is proposed for the communication of set partitioning in hierarchical trees image streams over wireless channels. The proposed scheme employs turbo codes and Reed-Solomon codes in order to deal effectively with burst errors. An algorithm for the optimal unequal error protection of the compressed bitstream is also proposed and applied in conjunction with an inherently more efficient technique for product code decoding. The resulting scheme is tested for the transmission of images over wireless channels. Experimental evaluation clearly demonstrates the superiority of the proposed transmission system in comparison to well-known robust coding schemes.

A new system is proposed for gait analysis and recognition applications. The new system is based on a denoising process and a new angular transform that are applied on binary silhouettes. Each human silhouette in a gait sequence is transformed into a low dimensional feature vector consisting of average pixel distances from the center of the silhouette. The sequence of feature vectors corresponding to a gait sequence is used for identification based on a minimum-distance criterion between test and reference sequences. By using the new system on the gait challenge database, improvements in recognition performance are seen in comparison to other methods of similar or higher complexity.

In this paper, a novel framework for multiple description coding of video is presented. The proposed scheme is based on a wavelet source coder and an efficient methodology for the generation of multiple descriptions. An algorithm is also presented for the optimal, in the sense of maximizing reconstruction quality, allocation of redundancy among the descriptions. Experimental results for the transmission of video using two descriptions demonstrate the efficiency of the proposed method.

In this paper, the case where a low-resolution image is initially watermarked and the watermark is afterwards detected on a noisy interpolated version of the watermarked image is investigated, using a correlation detector. Polyphase decomposition is utilized at the detector side in order to enable the flexible formation of a fused image, which is appropriate for watermark detection. The optimal fused correlator, obtained by combining information from different image components, is derived through a statistical analysis of the correlation detector properties and employment of Lagrange multipliers. It is shown that it is always preferable to perform detection on a fused image rather than the original image. Experimental results establish the efficiency of the proposed scheme.

A novel scheme is proposed for the transmission of JPEG2000 image streams over wireless channels. The proposed scheme exploits the block-based coding structure of the JPEG2000 streams and employs optimized product codes consisting of Turbo codes and Reed-Solomon codes in order to deal effectively with burst errors. The optimization is based on information extracted directly from the compressed JPEG2000 streams. Experimental evaluation demonstrates that the proposed scheme outperform other recent algorithms for the wireless transmission of images.

We propose a methodology for gait recognition based on dynamic time warping. The gait sequences are initially partitioned into gait cycles and then the test cycles are compared to reference cycles using dynamic time warping. The final distance between a test and a reference sequence is determined using a nonlinear rule. Experimental results are reported showing an improvement in recognition performance in comparison to the baseline algorithm on the "gait challenge" database.

In this paper, a novel algorithm is presented for the real-time, compressed-domain, unsupervised segmentation of image sequences and is applied to video indexing and retrieval. The segmentation algorithm uses motion and color information directly extracted from the MPEG-2 compressed stream. An iterative rejection scheme based on the bilinear motion model is used to effect foreground/background segmentation. Following that, meaningful foreground spatiotemporal objects are formed by initially examining the temporal consistency of the output of iterative rejection, clustering the resulting foreground macroblocks to connected regions and finally performing region tracking. Background segmentation to spatiotemporal objects is additionally performed. MPEG-7 compliant low-level descriptors describing the color, shape, position, and motion of the resulting spatiotemporal objects are extracted and are automatically mapped to appropriate intermediate-level descriptors forming a simple vocabulary termed object ontology. This, combined with a relevance feedback mechanism, allows the qualitative definition of the high-level concepts the user queries for (semantic objects, each represented by a keyword) and the retrieval of relevant video segments. Desired spatial and temporal relationships between the objects in multiple-keyword queries can also be expressed, using the shot ontology. Experimental results of the application of the segmentation algorithm to known sequences demonstrate the efficiency of the proposed segmentation approach. Sample queries reveal the potential of employing this segmentation algorithm as part of an object-based video indexing and retrieval scheme.

A novel image transmission scheme is proposed for the communication of SPIHT image streams over wireless channels. The proposed scheme employs turbo codes and erasure-correction codes in order to deal effectively with burst errors. An algorithm for the optimal unequal error protection of the compressed bitstream is also proposed. The resulting scheme is tested for the transmission of images over wireless channels. Experimental evaluation clearly demonstrates the superiority of the proposed scheme in comparison to well-known robust coding schemes.

A novel embedded wavelet coding scheme is proposed for the transmission of images over unreliable channels. The proposed scheme is based on the partitioning of information into a number of layers which can be decoded independently provided that some important and highly protected information is initially errorlessly transmitted to the decoder. Forward error correction is used in conjunction with the error-resilient source coder for the protection of the compressed stream. Unlike many other robust coding schemes presented to date, the proposed scheme is able to decode portions of the bitstream even after the occurrence of uncorrectable errors. This coding strategy is very suitable for application with block coding schemes such as defined by the JPEG2000 standard. The proposed scheme is compared with other robust image coders and is shown to be very suitable for transmission of images over memoryless channels.

An error-resilient coding scheme is proposed for the transmission of images over unreliable channels. Forward error correction is used in conjunction with the error-resilient source coder for the protection of the compressed stream. Unlike almost all other robust coding schemes presented to date, the proposed scheme is able to decode portions of the bitstream even after the occurrence of uncorrectable errors. The resulting coder is shown to be very efficient for image transmission over noisy channels.

The decoding of convolutional codes using the list Viterbi algorithm is proposed for data hiding applications. The performance of this technique is evaluated for wavelet-domain information hiding and is shown in many cases to be advantageous in comparison to the widely used turbo codes for efficient extraction of information embedded in digital images.

A new technique for watermarking of MPEG compressed video streams is proposed. The watermarking scheme operates directly in the domain of MPEG program streams. Perceptual models are used during the embedding process in order to preserve the video quality. The watermark is embedded in the compressed domain and is detected without the use of the original video sequence. Experimental evaluation demonstrates that the proposed scheme is able to withstand a variety of attacks. The resulting watermarking system is fast and reliable, and is suitable for copyright protection and real-time content authentication applications.

We propose novel algorithms for stereoscopic image coding based on the hierarchical decomposition of stereo information. The proposed schemes, based on the wavelet transform and zerotree quantization, are endowed with excellent progressive transmission capability and retain the option for perfect reconstruction of the original image pair. Experimental evaluation shows that the resulting methods produce superior results when compared with other algorithms for stereo image coding. This is achieved without introducing blocking artifacts and with the valuable additional convenience of the production of embedded bitstreams.

A method is presented for the optimization of arbitrary quantizers by use of a compensating postfilter. It is shown that the resulting optimized quantizers fit the model of a linear time-invariant filter followed by additive noise uncorrelated with the input which also characterizes the optimal (Lloyd-Max) quantizers. On the basis of this model, an expression for the variance of the error of a subband coder using optimized quantizers is explicitly determined. Given analysis filters which statistically separate the subbands, it is shown that this variance is minimized if these synthesis filters are chosen, which would achieve perfect reconstruction in lossless coding. The globally optimum filter bank, minimizing the coder error variance, is further obtained by proper choice of its analysis filters. A novel method for the determination of optimal bit allocation to subbands of the filter banks with optimized quantizers is also developed. The results are evaluated experimentally by comparison of the optimum uniformly split subband image coding scheme to classical logarithmically-split filter bank (wavelet) coding methods.

A novel framework for multiple description coding of video is presented. The proposed scheme is based on an efficient methodology for the generation of multiple descriptions which eliminates drift at the decoder. An arbitrary number of descriptions can be obtained from the same original video sequence. Experimental results for the transmission of video using two descriptions demonstrate the efficiency of the proposed methodology.

The optimal predictors of a lifting scheme in the general n-dimensional case are obtained and applied for the lossless compression of still images using first quincunx sampling and then simple row-column sampling. In each case, the efficiency of the linear predictors is enhanced nonlinearly. Directional postprocessing is used in the quincunx case, and adaptive-length postprocessing in the row-column case. Both methods are seen to perform well. The resulting nonlinear interpolation schemes achieve extremely efficient image decorrelation. We further investigate context modeling and adaptive arithmetic coding of wavelet coefficients in a lossless compression framework. Special attention is given to the modeling contexts and the adaptation of the arithmetic coder to the actual data. Experimental evaluation shows that the best of the resulting coders produces better results than other known algorithms for multiresolution-based lossless image coding.

We propose a novel algorithm for embedded stereoscopic image coding based on the hierarchical decomposition of stereo information. The proposed scheme is endowed with excellent progressive transmission capability and retains the option for perfect reconstruction of the original image pair. Experimental evaluation shows that the resulting method produces comparable results with a previously proposed algorithm for progressive stereo image coding.

In this paper we investigate the coding of 3D medical images using three-dimensional wavelet decomposition and adaptive arithmetic coding. A highly efficient context arithmetic coding scheme is used in conjunction with classical lifting-based wavelet decompositions. The resulting coder is shown to attain superior performance in comparison to other compression schemes for 3D medical image coding.

This paper presents a modified JPEG coder that is applied to the compression of mixed documents (containing text, natural images, and graphics) for printing purposes. The modified JPEG coder proposed in this paper takes advantage of the distinct perceptually significant regions in these documents to achieve higher perceptual quality than the standard JPEG coder. The region-adaptivity is performed via classified thresholding being totally compliant with the baseline standard. A computationally efficient classification algorithm is presented, and the improved performance of the classified JPEG coder is verified.

This correspondence introduces a novel method for the construction of extended reduced pyramids with rational decimation ratios from stage to stage. It is shown that this construction produces more accurate interpolation, and thus more efficient lossless compression, than conventional reduced pyramids.

We first calculate the optimal predictors of a lifting scheme in the general n-dimensional case. Then, we apply these optimal predictor filters with corresponding update filters for the lossless compression of still images using simple row-column sampling. The efficiency of the linear predictor is enhanced by nonlinear means, namely by adaptive-length postprocessing. The resulting filter bank in effect adapts to the features of the area of the image being processed and in this way it is shown to achieve superior performance than even the most efficient known lossless compression methods capable of progressive transmission.