The accurate segmentation of cells in cervical images is crucial for the recognition of pathological situations and the estimation of their severity. In this work, we investigate the segmentation of both the nucleus and the cytoplasm of each cell based on two Generative Adversarial Networks (GANs). First, we detect the location of the nucleus with the extraction of the nucleus boundaries in each cell, which is obtained by the training of the Nucleus-GAN. The segmented nucleus area serves as a guide factor for the definition of the cell boundary, and it is used as input in the Cell-GAN, for the segmentation of the cell boundaries. As it is verified by the experimental results, the proposed method is efficient and leads to accurate nucleus and cell boundaries, presenting high performance.

The work aims to design and implement a 3D interactive and addictive object avoidance game using the Unity platform. The implementation of the immersive virtual reality application uses any smart mobile device as an input and output device, utilizing its accelerometer and compass to record the orientation and rotation data of the device in 3D space and capture the digital environment stereoscopically on the device screen. A comparative study between a virtual reality and a desktop real-time 3D game is performed to analyze the various attributes of the game and determine which medium is most effective.

High-resolution images play an essential role in the performance of image analysis and pattern recognition methods. However, the expensive setup required to generate them and the inherent limitations of the sensors in optics manufacturing technology leads to the restricted availability of these images. In this work, we exploit the information retrieved in feature maps using the notable VGG networks and apply a transformer network to address spatial rigid affine transformation invariances, such as translation, scaling, and rotation. To evaluate and compare the performance of the model, three publicly available datasets were used. The model achieved very gratifying and accurate performance in terms of image PSNR and SSIM metrics against the baseline method.

In this work, we solve the problem of motion representation in videos, according to local transformations applied to specific keypoints extracted from static the images. First, we compute the coordinates of the keypoints of the body or face through a pre-trained model, and then we introduce a convolutional neural network to estimate a dense motion field through optical flow. Next, we train a generative adversarial network that exploits the previous information to generate new images that resemble as much as possible the target frames. To reduce trembling and extract smooth movements, our model incorporates a low-pass spatio-temporal Gaussian filter. Results indicate that our method provides high performance and the movement of objects is accurate and robust.

Augmented reality (AR) applications are regarded as effective experiential marketing practices that can help companies promote their products/services in an interactive manner and deliver exceptional consumer experiences. The purpose of the present study is to evaluate a wine-label AR mobile application by examining its impact on consumer experience dimensions, satisfaction, and re-usage intentions towards the application, as well as attitude and purchase intentions towards the wine product. Moreover, to test the effect of product-consumption related factors (consumption frequency, amount of spending, wine expertise, and attention to wine labels) and technology-related factors (consumers’ familiarity with smartphone applications, number of AR applications used in smartphone, and extent of information search for wine-related information through smartphones) on the experiential dimensions of entertainment, flow, escapism, and education. Towards this end, a wine AR label application was developed and evaluated using a quantitative survey. In total, 306 respondents answered a self-administered questionnaire after interacting with the application. Results indicate that the AR application induced the entertainment and educational dimensions of consumer experience. The AR experience was also able to increase respondents’ satisfaction with the application and in turn help them form positive attitudes and purchase intentions for the wine. Moreover, the present study revealed that respondents’ expertise for wine, attention to wine label, familiarity with smartphone applications, and information search for wine-related information through smartphones are important factors that have an impact on the experience lived by consumers when using the AR label application.

Multiphase materials are encountered in several areas of science and technology. Their properties are determined by the fraction of the phases (material compounds) constituting the composite material. Therefore, the quantitative characterization of phase fractions is highly demanded and has been the subject of extensive studies. To this end, a widely used technique is the segmentation of top-down back-scattered electron SEM (BSE-SEM) images given that different phases are depicted with pixel collections of different luminosity. Gaussian mixture models (GMM) are one the most popular and easily implemented methods to segment the BSE-SEM images through the deconvolution of their histograms. However, the accuracy and the limitations of their application have not been fully investigated. The aim of this paper is to design a neural-network approach to fill this gap and provide a fast tool for the automatic evaluation of the accuracy of GMM predictions for all material phases based on the inspection of the measured SEM image histogram alone. The proposed tool facilitates the decision-making process of an SEM user concerning the optimum choice of a segmentation method.

In this paper, we study the concepts, materials, tools, and applications that constitute what we call augmented reality (AR) for the wine industry. A comprehensive review of what are the basic multimedia content and the minimum algorithmic requirements used to implement successful AR applications for wine products is given. To this end, we provide a detailed analysis of how AR technology is used to create augmented “live” wine labels, and how digital storytelling has revolutionized wine products marketing. Also, we describe the use of AR technology to promote winemaking companies to influence consumer preferences. Finally, we report the characteristics of future research directions and some open issues and challenges on using AR for wine product promotion.

In this paper, the task of gender and age recognition is performed on pedestrian still images, which are usually captured in-the-wild with no near face-frontal information. Moreover, another difficulty originates from the underlying class imbalance in real examples, especially for the age estimation problem. The scope of the paper is to examine how different loss functions in convolutional neural networks (CNN) perform under the class imbalance problem. For this purpose, as a backbone, we employ the Residual Network (ResNet). On top of that, we attempt to benefit from appearance-based attributes, which are inherently present in the available data. We incorporate this knowledge in an autoencoder, which we attach to our baseline CNN for the combined model to jointly learn the features and increase the classification accuracy. Finally, all of our experiments are evaluated on two publicly available datasets.

Advanced methodologies for transmitting compressed images, within acceptable ranges of transmission rate and loss of information, make it possible to transmit a medical image through a communication channel. Most prior works on 3D medical image compression consider volumetric images as a whole but fail to account for the spatial and temporal coherence of adjacent slices. In this paper, we set out to develop a 3D medical image compression method that extends the 3D wavelet difference reduction algorithm by computing the similarity of the pixels in adjacent slices and progressively compress only the similar slices. The proposed method achieves high-efficiency performance on publicly available datasets of MRI scans by achieving compression down to one bit per voxel with PSNR and SSIM up to 52.3 dB and 0.7578, respectively.

Many approaches for action recognition focus on general actions, such as “running” or “walking”. This work presents a method for recognizing carrying actions in single images, by utilizing privileged information, such as annotations, available only during training, following the learning using privileged information paradigm. In addition, we introduce a dataset for carrying actions, formed using images extracted from YouTube videos depicting several scenarios. We accompany the dataset with a variety of different annotation types that include human pose, object and scene attributes. The experimental results demonstrate that our method, boosted sample averaged F1 score performance by 15.4% and 4.15% respectively, in the validation and testing partitions of our dataset, when compared to an end-to-end CNN model, trained only with observable information.

Hidden conditional random fields (HCRFs) are a powerful supervised classification system, which is able to capture the intrinsic motion patterns of a human action. However, finding the optimal number of hidden states remains a severe limitation for this model. This paper addresses this limitation by proposing a new model, called robust incremental hidden conditional random field (RI-HCRF). A hidden Markov model (HMM) is created for each observation paired with an action label and its parameters are defined by the potentials of the original HCRF graph. Starting from an initial number of hidden states and increasing their number incrementally, the Viterbi path is computed for each HMM. The method seeks for a sequence of hidden states, where each variable participates in a maximum number of optimal paths. Thereby, variables with low participation in optimal paths are rejected. In addition, a robust mixture of Student's t-distributions is imposed as a regularizer to the parameters of the model. The experimental results on human action recognition show that RI-HCRF successfully estimates the number of hidden states and outperforms all state-of-the-art models.

An algorithm for the localization and counting of cells in histopathological images is presented. The algorithm relies on the presegmentation of an image into a number of superpixels followed by two random forests for classification. The first random forest determines if there are any cells in the superpixels at its input and the second random forest provides the number of cells in the respective superpixel. The algorithm is evaluated on a bone marrow histopathological dataset. We argue that a single random forest is not sufficient to detect all the cells in the image while a cascade of classifiers achieves higher accuracy. The results compare favorably with the state of the art but with a lower computational cost.

Classification models may often suffer from “structure imbalance” between training and testing data that may occur due to the deficient data collection process. This imbalance can be represented by the learning using privileged information (LUPI) paradigm. In this paper, we present a supervised probabilistic classification approach that integrates LUPI into a hidden conditional random field (HCRF) model. The proposed model is called LUPI-HCRF and is able to cope with additional information that is only available during training. Moreover, the proposed method employes Student's t-distribution to provide robustness to outliers by modeling the conditional distribution of the privileged information. Experimental results in three publicly available datasets demonstrate the effectiveness of the proposed approach and improve the state-of-the-art in the LUPI framework for recognizing human activities.

Incorporating additional knowledge in the learning process can be beneficial for several computer vision and machine learning tasks. Whether privileged information originates from a source domain that is adapted to a target domain, or as additional features available at training time only, using such privileged (i.e., auxiliary) information is of high importance as it improves the recognition performance and generalization. However, both primary and privileged information are rarely derived from the same distribution, which poses an additional challenge to the recognition task. To address these challenges, we present a novel learning paradigm that leverages privileged information in a domain adaptation setup to perform visual recognition tasks. The proposed framework, named Adaptive SVM+, combines the advantages of both the learning using privileged information (LUPI) paradigm and the domain adaptation framework, which are naturally embedded in the objective function of a regular SVM. We demonstrate the effectiveness of our approach on the publicly available Animals with Attributes and INTERACT datasets and report state-of-the-art results in both of them.

In many human activity recognition systems the size of the unlabeled training data may be significantly large due to expensive human effort required for data annotation. Moreover, the insufficient data collection process from heterogenous sources may cause dissimilarities between training and testing data. To address these limitations, a novel probabilistic approach that combines learning using privileged information (LUPI) and active learning is proposed. A pool-based privileged active learning approach is presented for semi-supervising learning of human activities from multimodal labeled and unlabeled data. Both uncertainty and distance from the decision boundary are used as a query inference strategies for selecting an unlabeled observation and query its label. Experimental results in four publicly available datasets demonstrate that the proposed method can identify with high accuracy complex human activities.

Most of the facial expression recognition methods consider that both training and testing data are equally distributed. As facial image sequences may contain information for heterogeneous sources, facial data may be asymmetrically distributed between training and testing, as it may be difficult to maintain the same quality and quantity of information. In this work, we present a novel classification method based on the learning using privileged information (LUPI) paradigm to address the problem of facial expression recognition. We introduce a probabilistic classification approach based on conditional random fields (CRFs) to indirectly propagate knowledge from privileged to regular feature space. Each feature space owns specific parameter settings, which are combined together through a Gaussian prior, to train the proposed t-CRF+ model and allow the different tasks to share parameters and improve classification performance. The proposed method is validated on two challenging and publicly available benchmarks on facial expression recognition and improved the state-of-the-art methods in the LUPI framework.

The automated interpretation of Pap smear images is a challenging issue with several aspects. The accurate segmentation of the structuring elements of each cell is a crucial procedure which entails in the correct identification of pathological situations. However, the extended cell overlapping in Pap smear slides complicates the automated analysis of these cytological images. In this work, we propose an efficient algorithm for the separation of the cytoplasm area of overlapping cells. The proposed method is based on the fact that in isolated cells the pixels of the cytoplasm exhibit similar features and the cytoplasm area is homogeneous. Thus, the observation of intensity changes in extended subareas of the cytoplasm indicates the existence of overlapping cells. In the first step of the proposed method, the image is tesselated into perceptually meaningful individual regions using a superpixel algorithm. In a second step, these areas are merged into regions exhibiting the same characteristics, resulting in the identification of each cytoplasm area and the corresponding nuclei. The area of overlap is then detected using an algorithm that specifies faint changes in the intensity of the cytoplasm of each cell. The method has been evaluated on cytological images of conventional Pap smears, and the results are very promising.

A human behavior recognition method with an application to political speech videos is presented. We focus on modeling the behavior of a subject with a conditional random field (CRF). The unary terms of the CRF employ spatiotemporal features (i.e., HOG3D, STIP and LBP). The pairwise terms are based on kinematic features such as the velocity and the acceleration of the subject. As an exact solution to the maximization of the posterior probability of the labels is generally intractable, loopy belief propagation was employed as an approximate inference method. To evaluate the performance of the model, we also introduce a novel behavior dataset, which includes low resolution video sequences depicting different people speaking in the Greek parliament. The subjects of the Parliament dataset are labeled as friendly, aggressive or neutral depending on the intensity of their political speech. The discrimination between friendly and aggressive labels is not straightforward in political speeches as the subjects perform similar movements in both cases. Experimental results show that the model can reach high accuracy in this relatively difficult dataset.

A framework for action representation and recognition based on the description of an action by time series of optical flow motion features is presented. In the learning step, the motion curves representing each action are clustered using Gaussian mixture modeling (GMM). In the recognition step, the optical flow curves of a probe sequence are also clustered using a GMM and the probe curves are matched to the learned curves using a non-metric similarity function based on the longest common subsequence which is robust to noise and provides an intuitive notion of similarity between trajectories. Finally, the probe sequence is categorized to the learned action with the maximum similarity using a nearest neighbor classification scheme. Experimental results on common action databases demonstrate the effectiveness of the proposed method.

In this work, we propose an adaptive M-estimation scheme for robust image super-resolution. The proposed algorithm relies on a maximum a posteriori (MAP) framework and addresses the presence of outliers in the low resolution images. Moreover, apart from the robust estimation of the high resolution image, the contribution of the method is twofold: (i) the robust computation of the regularization parameters controlling the relative strength of the prior with respect to the data fidelity term and (ii) the robust estimation of the optimal step size in the update of the high resolution image. Experimental results demonstrate that integrating these estimations into a robust framework leads to significant improvement in the accuracy of the high resolution image.

Accurate image registration plays a preponderant role in image super-resolution methods and in the related literature landmarkbased registration methods have gained increasing acceptance in this framework. However, their solution relies on point correspondences and on least squares estimation of the registration parameters necessitating further improvement. In this work, a maximum a posteriori scheme for image super-resolution is presented where the image registration part is accomplished in two steps. At first, the lowresolution images are registered by establishing correspondences between robust SIFT features. In the second step, the estimation of the registration parameters is fine-tuned along with the estimation of the high resolution image, in an iterative scheme, using the maximization of the mutual information criterion. Numerical results showed that the reconstructed image is consistently of higher quality than in standard MAP-based methods employing only landmarks.