Fundamenta Informaticae - Volume 168, issue 2-4

Authors: Skarbek, Wladyslaw | Zhang, Yu-Dong

Article Type: Other

DOI: 10.3233/FI-2019-1826

Citation: Fundamenta Informaticae, vol. 168, no. 2-4, pp. i-ii, 2019

Authors: Skarbek, Władysław

Article Type: Research Article

Abstract: This tutorial material on Convolutional Neural Networks (CNN) and its applications in digital media research is based on the concept of Symbolic Tensor Neural Networks. The set of STNN expressions is specified in Backus-Naur Form (BNF) which is annotated by constraints typical for labeled acyclic directed graphs (DAG). The BNF induction begins from a collection of neural unit symbols with extra (up to five) decoration fields (including tensor depth and sharing fields). The inductive rules provide not only the general graph structure but also the specific shortcuts for residual blocks of units. A syntactic mechanism for network fragments modularization is …introduced via user defined units and their instances. Moreover, the dual BNF rules are specified in order to generate the Dual Symbolic Tensor Neural Network (DSTNN). The joined interpretation of STNN and DSTNN provides the correct flow of gradient tensors, back propagated at the training stage. The proposed symbolic representation of CNNs is illustrated for six generic digital media applications (CREAMS): Compression, Recognition, Embedding, Annotation, 3D Modeling for human-computer interfacing, and data Security based on digital media objects. In order to make the CNN description and its gradient flow complete, for all presented applications, the symbolic representations of mathematically defined loss/gain functions and gradient flow equations for all used core units, are given. The tutorial is to convince the reader that STNN is not only a convenient symbolic notation for public presentations of CNN based solutions for CREAMS problems but also that it is a design blueprint with a potential for automatic generation of application source code. Show more

Keywords: convolutional neural network, tensor neural network, deep learning, deep digital media application

DOI: 10.3233/FI-2019-1827

Citation: Fundamenta Informaticae, vol. 168, no. 2-4, pp. 89-184, 2019

Authors: Biedrzycki, Rafał | Zawistowski, Paweł | Twardowski, Bartłomiej

Article Type: Research Article

Abstract: In this paper we identify and formulate two optimization tasks solved in connection with training DL models and constructing adversarial examples. This guides our review of optimization methods commonly used within the DL community. Simultaneously, we present findings from the literature concerning metaheuristics and black-box optimization. We focus on well-known optimizers suitable for solving ℝN tasks, which achieve good results on benchmarks and in competitions. Finally, we look into the research connected with utilizing metaheuristic optimization methods in combination with deep learning models.

Keywords: Deep learning, metaheuristics, optimization

DOI: 10.3233/FI-2019-1828

Citation: Fundamenta Informaticae, vol. 168, no. 2-4, pp. 185-218, 2019

Authors: Yu, Xiang | Wang, Shui-Hua

Article Type: Research Article

Abstract: Breast cancer is one of the common cancers threatening the health of women while the incident rate of it is quite low in men to contribute to a major killer of men. Early syndromes of breast cancer including micro-calcification, mass, and distortion in mammography images can be very helpful for radiologists to make diagnosis of the cancer at early stage, which means the cancer can be treated or even be cured timely and thus make early diagnosis important. To assist radiologists with diagnosis, we set up a computer-aided diagnosis system to make diagnosis decision of breast cancer in this paper. …We acquired regions of interests in mammographic images from public database, and labeled regions containing micro-calcification or mass as abnormality while regions without such abnormalities as normality. By transferring the state-of-the-art networks into our quest, we found that ResNet18 performed best and achieved mean accuracy of 95.91%. Show more

Keywords: Abnormality, Diagnosis system, Transfer learning

DOI: 10.3233/FI-2019-1829

Citation: Fundamenta Informaticae, vol. 168, no. 2-4, pp. 219-230, 2019

Authors: Hong, Jin | Cheng, Hong | Wang, Shui-Hua | Liu, Jie

Article Type: Research Article

Abstract: The existence and distribution pattern of cerebral microbleeds (CMBs) are associated with some underlying aetiologies caused by intra-cerebral hemorrhage (ICH). CMBs as a kind of subclinical sign can be recognized via magnetic resonance (MR) imaging technique in a few years before the onset of the disease. Hence, detecting CMBs accurately is important for treating and preventing related cerebral disease. In this study, we employed convolution neural network (CNN) for CMBs detection because of its powerful ability in image recognition. In view of too many efforts on optimizing the structure of CNN for achieving a better performance, we introduced center loss, …which can greatly enhance the discriminative power of the deeply learned features, to CMBs detection for the first time. It is found that the performances of convolution neural network (CNN) trained under the joint supervision of softmax loss and center loss were significantly better than that under the supervision of softmax loss, even if there are few mislabelled samples in training data. With this trick, we achieved a high performance with a sensitivity of 98.869 ± 1.026%, a specificity of 96.491 ± 0.367%, and an accuracy of 97.681 ± 0.497%, which is better than four state-of-the-art methods. Show more

Keywords: cerebral microbleeds, convolution neural network, discriminative feature learning, center loss

DOI: 10.3233/FI-2019-1830

Citation: Fundamenta Informaticae, vol. 168, no. 2-4, pp. 231-248, 2019

Authors: Kurowski, Adam | Mrozik, Katarzyna | Kostek, Bozena | Czyzewski, Andrzej

Article Type: Research Article

Abstract: A method for assessing separability of EEG signals associated with three classes of brain activity is proposed. The EEG signals are acquired from 23 subjects, gathered from a headset consisting of 14 electrodes. Data are processed by applying Discrete Wavelet Transform (DWT) for the signal analysis and an autoencoder neural network for the brain activity separation. Processing involves 74 wavelets from 3 DWT families: Coiflets, Daubechies and Symlets. Euclidean distance between clusters normalized with respect to the standard deviation of the whole set of data are used to separate each task performed by participants. The results of this stage allow …for an assessment of separability between subsets of data associated with each activity performed by experiment participants. The speed of convergence of the training process employing deep learning-based clustering is also measured. Show more

Keywords: brain activity, brain-computer interfaces, EEG, data clustering, machine learning, deep learning

DOI: 10.3233/FI-2019-1831

Citation: Fundamenta Informaticae, vol. 168, no. 2-4, pp. 249-268, 2019

Authors: Tautkutė, Ivona | Trzciński, Tomasz

Article Type: Research Article

Abstract: Classification of human emotions remains an important and challenging task for many computer vision algorithms, especially in the era of humanoid robots which coexist with humans in their everyday life. Currently proposed methods for emotion recognition solve this task using multi-layered convolutional networks that do not explicitly infer any facial features in the classification phase. In this work, we postulate a fundamentally different approach to solve emotion recognition task that relies on incorporating facial landmarks as a part of the classification loss function. To that end, we extend a recently proposed Deep Alignment Network (DAN) with a term related to …facial features. Thanks to this simple modification, our model called EmotionalDAN is able to outperform state-of-the-art emotion classification methods on two challenging benchmark dataset by up to 5%. Furthermore, we visualize image regions analyzed by the network when making a decision and the results indicate that our EmotionalDAN model is able to correctly identify facial landmarks responsible for expressing the emotions. Show more

Keywords: machine learning, emotion recognition, facial expression recognition

DOI: 10.3233/FI-2019-1832

Citation: Fundamenta Informaticae, vol. 168, no. 2-4, pp. 269-285, 2019

Authors: Pilarczyk, Rafał | Chang, Xin | Skarbek, Władysław

Article Type: Research Article

Abstract: Several computer algorithms for recognition of visible human emotions are compared at the web camera scenario using CNN/MMOD face detector. The recognition refers to four face expressions: smile, surprise, anger, and neutral. At the feature extraction stage, the following three concepts of face description are confronted: (a) static 2D face geometry represented by its 68 characteristic landmarks (FP68); (b) dynamic 3D geometry defined by motion parameters for eight distinguished face parts (denoted as AU8) of personalized Candide-3 model; (c) static 2D visual description as 2D array of gray scale pixels (known as facial raw image). At the classification stage, the …performance of two major models are analyzed: (a) support vector machine (SVM) with kernel options; (b) convolutional neural network (CNN) with variety of relevant tensor processing layers and blocks of them. The models are trained for frontal views of human faces while they are tested for arbitrary head poses. For geometric features, the success rate (accuracy) indicate nearly triple increase of performance of CNN with respect to SVM classifiers. For raw images, CNN outperforms in accuracy its best geometric counterpart (AU/CNN) by about 30 percent while the best SVM solutions are inferior. For F-score the high advantage of raw/CNN over geometric/CNN and geometric/SVM is observed, as well. We conclude that contrary to CNN based emotion classifiers, the generalization capability wrt human head pose for SVM based emotion classifiers, is worse too. Show more

Keywords: face expression recognition, face landmarks, facial action units, SVM classifier, CNN classifier

DOI: 10.3233/FI-2019-1833

Citation: Fundamenta Informaticae, vol. 168, no. 2-4, pp. 287-310, 2019

Authors: Pęśko, Maciej | Svystun, Adam | Andruszkiewicz, Paweł | Rokita, Przemysław | Trzciński, Tomasz

Article Type: Research Article

Abstract: In this paper, we propose a solution to transform a video into a comics. We approach this task using a neural style algorithm based on Generative Adversarial Networks (GANs). Several recent works in the field of Neural Style Transfer showed that producing an image in the style of another image is feasible. In this paper, we build up on these works and extend the existing set of style transfer use cases with a working application of video comixification. To that end, we train an end-to-end solution that transforms input video into a comics in two stages. In the first stage, …we propose a state-of-the-art keyframes extraction algorithm that selects a subset of frames from the video to provide the most comprehensive video context and we filter those frames using image aesthetic estimation engine. In the second stage, the style of selected keyframes is transferred into a comics. To provide the most aesthetically compelling results, we selected the most state-of-the art style transfer solution and based on that implement our own ComixGAN framework. The final contribution of our work is a Web-based working application of video comixification available at http://comixify.ii.pw.edu.pl . Show more

Keywords: Neural Style Transfer, Style Transfer, Comics Style, Comics, Computer Vision, Neural Network

DOI: 10.3233/FI-2019-1834

Citation: Fundamenta Informaticae, vol. 168, no. 2-4, pp. 311-333, 2019

Authors: Sharma, Shailza | Bawa, Vivek Singh | Kumar, Vinay

Article Type: Research Article

Abstract: Image super resolution has gained a lot of attention due to its applications in different fields of image processing. It is used to produce high-resolution images from low-resolution input. Because of the excellent learning capability of convolution neural networks, these networks are able to learn complex spatial structures for image super-resolution. In this paper, two different architectures have been proposed for image super resolution. The first architecture is Dual Subpixel Layer Convolution Neural Network (DSL-CNN), which stacks two subpixel CNN architectures to enhance model depth for better representational capability. Two stages provide an effective upscaling factor of 4. In the …second architecture, named as Residue based Dual Subpixel Layer Convolution Neural Network (RDSL-CNN), two-stage residual learning has been introduced which effectively sustains the high frequency details and provides superior results than the previous state-of-the-art methods. The performance of the two architectures has been evaluated on various image datasets, and compared with other state-of-the-art methods. Show more

Keywords: Super resolution, convolutional neural network, residual learning, deep learning, subpixel layer

DOI: 10.3233/FI-2019-1835

Citation: Fundamenta Informaticae, vol. 168, no. 2-4, pp. 335-351, 2019