Today many companies exist and market their products and services on social medias, and therefore may receive reviews and thoughts from their end-users directly in these social medias. Reading every text by hand can be time-consuming, so by analysing the sentiment for all texts give the companies an overview how positive or negative the users are on a specific subject. Sentiment analysis is a feature that Beanloop AB is interested in implementing in their future projects and this thesis research problem was to investigate how deep learning could be used for this task. It was done by conducting an experiment with deep learning and neural networks. Several convolutional neural network models were implemented with different settings to find a combination of settings that gave the highest accuracy on the given test dataset. There were two different kind of models, one kind classifying positive and negative, and the second classified the previous two categories but also neutral. The training dataset and the test dataset contained data from two recommendation sites, www.reco.se and se.trustpilot.com. The final result shows that when classifying three categories (positive, negative and neutral) the models had problems to reach an accuracy at 85%, were only one model reached 80% accuracy as best on the test dataset. However, when only classifying two categories (positive and negative) the models showed very good results and reached almost 95% accuracy for every model.

Pedestrian detection is an important field with applications in active safety systems for cars as well as autonomous driving. Since autonomous driving and active safety are becoming technically feasible now the interest for these applications has dramatically increased.The aim of this thesis is to investigate convolutional neural networks (CNN) for pedestrian detection. The reason for this is that CNN have recently beensuccessfully applied to several different computer vision problems. The main applications of pedestrian detection are in real time systems. For this reason,this thesis investigates strategies for reducing the computational complexity offorward propagation for CNN.The approach used in this thesis for extracting pedestrians is to use a CNN tofind a probability map of where pedestrians are located. From this probabilitymap bounding boxes for pedestrians are generated. A method for handling scale invariance for the objects of interest has also been developed in this thesis. Experiments show that using this method givessignificantly better results for the problem of pedestrian detection.The accuracy which this thesis has managed to achieve is similar to the accuracy for some other works which use CNN.

We tackle the task of semi-supervised video object segmentation, i.e, pixel-level object classification of the images in video sequences using very limited ground truth training data of its corresponding video. Recently introduced online adaptation of convolutional neural networks for video object segmentation (OnAVOS) has achieved good results by pretraining the network, fine-tuning on the first frame and training the network at test time using its approximate prediction as newly obtained ground truth. We propose Flow Adaptive Video Object Segmentation (FAVOS) that refines the generated adaptive ground truth for online updates and utilizes temporal consistency between video frames with the help of optical flow. We validate our approach on the DAVIS Challenge and achieve rank 1 results on the DAVIS 2016 Challenge (single-object segmentation) and competitive scores on both DAVIS 2018 Semi-supervised Challenge and Interactive Challenge (multi-object segmentation). While most models tend to have increasing complexity for the challenging task of video object segmentation, FAVOS provides a simple and efficient pipeline that produces accurate predictions.

The diploma thesis deals with current problems of person identification and deep learning. Furthermore, the work deals mainly with obtaining quality and diverse data that are used to train deep learning with convolutional neural networks for face recognition. There is very little public access to such data, so the practical part focuses on creating the MakeHuman plugin that will generate a database of random face images. It is possible to generate faces according to five different scenarios in which purely random faces or faces where the same can be seen with modifications such as different hair, beard, hat, glasses and more are created. The scenarios also allow you to generate faces with some expressions or faces as they age. You can set some parameters that give the appearance of the resulting database in the plugin. This can include face images from different angles of rotation, zooming and lighting.

This diploma thesis deals with the kidney tumor segmentation in CT data. First kidney anatomy and pathology is discussed. Following topics are the conventional segmentation techniques and segmentation techniques using machine learning. In the final part, the convolutional neural network is discussed as its algoritm was used for segmentation in the practical part, in which algoritm for segmentation was designed in Python programming language. This algoritm was tested and evaluated using databaze KiTS19.

The thesis deals with problems of automatic detection of premature ventricular contractions in ECG records. One detection method which uses a convolutional neural network and LSTM units is implemented in the Python language. Cardiac cycles extracted from one-lead ECG were used for detection. F1 score for binary classification (PVC and normal beat) on the test dataset reached 96,41 % and 81,76 % for three-class classification (PVC, normal beat and other arrhythmias). Lastly, the accuracy of the classification is evaluated and discussed, the achieved results for binary classification are comparable to the results of methods described in different papers.

Machine translation (MT) as a field of research has known significant advances in recent years, with the increased interest for neural machine translation (NMT). By combining deep learning with translation, researchers have been able to deliver systems that perform better than most, if not all, of their predecessors. While the general consensus regarding NMT is that it renders higher-quality translations that are overall more idiomatic, researchers recognize that NMT systems still struggle to deal with certain classic difficulties, and that their performance may vary depending on their architecture. In this project, we implement a challenge-set based approach to the evaluation of examples of three main NMT architectures: convolutional neural network-based systems (CNN), recurrent neural network-based (RNN) systems, and attention-based systems, trained on the same data set for English to French translation. The challenge set focuses on a selection of lexical and syntactic difficulties (e.g., ambiguities) drawn from literature on human translation, machine translation, and writing for translation, and also includes variations in sentence lengths and structures that are recognized as sources of difficulties even for NMT systems. This set allows us to evaluate performance in multiple areas of difficulty for the systems overall, as well as to evaluate any differences between architectures’ performance. Through our challenge set, we found that our CNN-based system tends to reword sentences, sometimes shifting their meaning, while our RNN-based system seems to perform better when provided with a larger context, and our attention-based system seems to struggle the longer a sentence becomes.

With the evolution of machine learning algorithms they are seeing a wider use in traditional signal processing applications. One of these areas is in radios for improved signal identification algorithms. With the large computational complexity of convolutional neural networks, it is of importance to use platforms that are as fast and energy efficient as possible. This thesis investigates hardware acceleration of convolutional neural networks on field programmable gate arrays, an reconfigurable integrated circuit. An existing toolflow, Haddoc2, is used and evaluated. This tool automates the mapping of a convolutional neural network from a high level description in Caffe to a synthesisable hardware description in VHDL hardware description language. Four models of different sizes are trained on the MNIST dataset and accelerators for these at different bitwidths are generated and then simulated in a VHDL testbench. The resulting accuracies are tolerable for the target problem and Haddoc2 can produce fast accelerators that would work well for smaller networks. Big networks was found to consume large amounts of resources in the field programmable gate array and is not feasible for a practical application. The treatment of weights as constants makes the accelerators fast since there is no memory bottleneck but makes the accelerator less flexible since a new set of weights would require to re-synthesize the design and reprogramming the field programmable gate array.

In computer vision, it has in recent years become more popular to use point clouds to represent 3D data. To understand what a point cloud contains, methods like semantic segmentation can be used. Semantic segmentation is the problem of segmenting images or point clouds and understanding what the different segments are. An application for semantic segmentation of point clouds are e.g. autonomous driving, where the car needs information about objects in its surrounding. Our approach to the problem, is to project the point clouds into 2D virtual images using the Katz projection. Then we use pre-trained convolutional neural networks to semantically segment the images. To get the semantically segmented point clouds, we project back the scores from the segmentation into the point cloud. Our approach is evaluated on the semantic3D dataset. We find our method is comparable to state-of-the-art, without any fine-tuning on the Semantic3Ddataset.

Background. Since historical handwritten documents have played important roles in promoting the development of human civilization, many of them have been preserved through digital versions for more scientific researches. However, various degradations always exist in these documents, which could interfere in normal reading. But, binarized versions can keep meaningful contents without degradations from original document images. Document image binarization always works as a pre-processing step before complex document analysis and recognition. It aims to extract texts from a document image. A desirable binarization performance can promote subsequent processing steps positively. For getting better performance for document image binarization, efficient binarization methods are needed. In recent years, machine learning centered on deep learning has gathered substantial attention in document image binarization, for example, Convolutional Neural Networks (CNNs) are widely applied in document image binarization because of the powerful ability of feature extraction and classification. Meanwhile, Support Vector Machine (SVM) is also used in image binarization. Its objective is to build an optimal hyperplane that could maximize the margin between negative samples and positive samples, which can separate the foreground pixels and the background pixels of the image distinctly. Objectives. This thesis aims to explore how the CNN based process of deep convolutional feature extraction and an SVM classifier can be integrated well to binarize handwritten document images, and how the results are, compared with some state-of-the-art document binarization methods. Methods. To investigate the effect of the proposed method on document image binarization, it is implemented and trained. In the architecture, CNN is used to extract features from input images, afterwards these features are fed into SVM for classification. The model is trained and tested with six different datasets. Then, there is a performance comparison between the proposed model and other binarization methods, including some state-of-the-art methods on other three different datasets. Results. The performance results indicate that the proposed model not only can work well but also perform better than some other novel handwritten document binarization method. Especially, evaluation of the results on DIBCO 2013 dataset indicates that our method fully outperforms other chosen binarization methods on all the four evaluation metrics. Besides, it also has the ability to deal with some degradations, which demonstrates its generalization and learning ability are excellent. When a new kind of degradation appears, the proposed method can address it properly even though it never appears in the training datasets. Conclusions. This thesis concludes that the CNN based component and SVM can be combined together for handwritten document binarization. Additionally, in certain datasets, it outperforms some other state-of-the-art binarization methods. Meanwhile, its generalization and learning ability is outstanding when dealing with some degradations.