For centuries physicians, scientists, engineers, mathematicians, and many others have been asking: 'what are the forces that drive tissues in an embryo to their final geometric forms?' At the tissue and whole embryo level, a multitude of very different morphogenetic processes, such as gastrulation and neurulation are involved. However, at the cellular level, virtually all of these processes are evidently driven by a relatively small number of internal structures all of whose forces can be resolved into equivalent interfacial tensions γ. Measuring the cell-level forces that drive specific morphogenetic events remains one of the great unsolved problems of biomechanics. Here I present a novel approach that allows these forces to be estimated from time lapse images. In this approach, the motions of all visible triple junctions formed between trios of cells adjacent to each other in epithelia (2D cell sheets) are tracked in time-lapse images. An existing cell-based Finite Element (FE) model is then used to calculate the viscous forces needed to deform each cell in the observed way. A recursive least squares technique with variable forgetting factors is then used to estimate the interfacial tensions that would have to be present along each cell-cell interface to provide those forces, along with the attendant pressures in each cell. The algorithm is tested extensively using synthetic data from an FE model. Emphasis is placed on features likely to be encountered in data from live tissues during morphogenesis and wound healing. Those features include algorithm stability and tracking despite input noise, interfacial tensions that could change slowly or suddenly, and complications from imaging small regions of a larger epithelial tissue (the frayed boundary problem). Although the basic algorithm is highly sensitive to input noise due to the ill-conditioned nature of the system of equations that must be solved to obtain the interfacial tensions, methods are introduced to improve the resulting force and pressure estimates. The final algorithm returns very good estimates for interfacial tensions and intracellular cellular pressures when used with synthetic data, and it holds great promise for calculating the forces that remodel live tissue.

The diploma thesis deals with synthetic data, selected approaches to their generation together with a practical task of data generation. The goal of the thesis is to describe the selected approaches to data generation, capture their key advantages and disadvantages and compare the individual approaches to each other. The practical part of the thesis describes generation of synthetic data for teaching knowledge discovery using databases. The thesis includes a basic description of synthetic data and thoroughly explains the process of their generation. The approaches selected for further examination are random data generation, the statistical approach, data generation languages and the ReverseMiner tool. The thesis also describes the practical usage of synthetic data and the suitability of each approach for certain purposes. Within this thesis, educational data Hotel SD were created using the ReverseMiner tool. The data contain relations discoverable with SD (set-difference) GUHA-procedures.

With the increasing demand for labeled data in machine learning for visual perception tasks, the interest in using synthetically generated data has grown. Due to the existence of a domain gap between synthetic and real data, strategies in domain adaptation are necessary to achieve high performance with models trained on synthetic or mixed data. With a dataset of synthetically blocked fish-eye lenses in traffic environments, we explore different strategies to train a neural network. The neural network is a binary classifier for full blockage detection. The different strategies tested are data mixing, fine-tuning, domain adversarial training, and adversarial discriminative domain adaptation. Different ratios between synthetically generated data and real data are also tested. Our experiments showed that fine-tuning had slightly superior results in this test environment. To fully take advantage of the domain adversarial training, training until domain indiscriminate features are learned is necessary and helps the model attain higher performance than using random data mixing.

Good generative models for short time series data exist and have been applied for both data augmentation and privacy protection purposes in the past. A common theme for existing generative models is that they all use a recurrent neural network (RNN) architecture, which makes the models limited regarding the length of the sequences. In real world problems, we might have to deal with data containing longer sequences, and it is such data we in this thesis attempt to synthesize. By combining the recently successful TimeGAN framework with a temporal convolutional network component architecture, we generate synthetic sequential data for two toy data sets: sequential MNIST and multivariate sine waves. The results strongly indicate, although relying solely on a visual inspection, that the model manage to capture long temporal dynamics over time and also relations between different features for the multivariate sine waves data set. In order to make our model applicable for real world data sets, we suggest two improvements. Firstly, the validation of the generated data should not only rely on visual inspection, but also ensure that the synthetic data has the same statistical distribution. Secondly, depending on the task, model refinements such that the synthetic samples look even more realistic should be made.

Autonomous vehicles perform semantic segmentation to orient themselves, but training neural networks for semantic segmentation requires large amounts of labeled data. A hand-labeled real-life dataset requires considerable effort to create, so we instead turn to virtual simulators where the segmented labels are known to generate large datasets virtually for free. This work investigates how effective synthetic datasets are in driving scenarios by collecting a dataset from a simulator and testing it against a real-life hand-labeled dataset. We show that we can get a model up and running faster by mixing synthetic and real-life data than traditional dataset collection methods and achieve close to baseline performance.

Modern camera-based vehicle safety systems heavily rely on machine learning and consequently require large amounts of training data to perform reliably. However, collecting and annotating the needed data is an extremely expensive and time-consuming process. In addition, it is exceptionally difficult to collect data that covers critical scenarios. This thesis investigates to what extent synthetic data can replace real-world data for these scenarios. Since only a limited amount of data consisting of such real-world scenarios is available, this thesis instead makes use of proxy scenarios, e.g. situations when pedestrians are located closely in front of the vehicle (for example at a crosswalk). The presented approach involves training a detector on real-world data where all samples of these proxy scenarios have been removed and compare it to other detectors trained on data where the removed samples have been replaced with various degrees of synthetic data. A method for generating and automatically and accurately annotating synthetic data, using features in the CARLA simulator, is presented. Also, the domain gap between the synthetic and real-world data is analyzed and methods in domain adaptation and data augmentation are reviewed. The presented experiments show that aligning statistical properties between the synthetic and real-world datasets distinctly mitigates the domain gap. There are also clear indications that synthetic data can help detect pedestrians in critical traffic situations / <p>Examensarbetet är utfört vid Institutionen för teknik och naturvetenskap (ITN) vid Tekniska fakulteten, Linköpings universitet</p>

Mastitis is the most common disease affecting Swedish milk cows. Automatic image classification can be useful for quickly classifying the bacteria causing this inflammation, in turn making it possible to start treatment more quickly. However, training an automatic classifier relies on the availability of data. Data collection can be a slow process, and GANs are a promising way to generate synthetic data to add plausible samples to an existing data set. The purpose of this thesis is to explore the usefulness of GANs for generating images of bacteria. This was done through researching existing literature on the subject, implementing a GAN, and evaluating the generated images. A cGAN capable of generating class-specific bacteria was implemented and improvements upon it made. The images generated by the cGAN were evaluated using visual examination, rapid scene categorization, and an expert interview regarding the generated images. While the cGAN was able to replicate certain features in the real images, it fails in crucial aspects such as symmetry and detail. It is possible that other GAN variants may be better suited to the task. Lastly, the results highlight the challenges of evaluating GANs with current evaluation methods.

The lack of legitimate datasets on mobile money transactions toperform research on in the domain of fraud detection is a big prob-lem today in the scientic community. Part of the problem is theintrinsic private nature of mobile transactions, not much infor-mation can be exploited. This will leave the researchers with theburden of rst harnessing the dataset before performing the actualresearch on it. The dataset corresponds to the set of data in whichthe research is to be performed on. This thesis discusses a solutionto such a problem, namely the Paysim simulator. Paysim is a -nancial simulator that simulates mobile money transactions basedon an original dataset. We present a solution to ultimately yieldthe possibility to simulate mobile money transactions in such a waythat they become similar to the original dataset. The similarity orthe congruity will be measured by calculating the error-rate betweenthe synthetic data set and the original data set. With technologyframeworks such as "Agent Based" simulation techniques, and theapplication of mathematical statistics, it can be demonstrated thatthe synthetic data is as prudent as the original data set. The aimof this thesis is to demonstrate with statistical models that PaySimcan be used as a tool for the intents of nancial simulations.

The task of 6D pose estimation with deep learning is to train networks to, from an im-age of an object, determine the rotation and translation of the object. Impressive resultshave recently been shown in deep learning based 6D pose estimation. However, many cur-rent solutions rely on real-world data when training, which as opposed to synthetic data,requires time consuming annotation. In this thesis, we introduce a pipeline for generatingsynthetic ground truth data for deep 6D pose estimation, where annotation is done auto-matically. With a 3D CAD-model, we use Blender to render 2D images of the model fromdifferent view points. We also create all other relevant data needed for pose estimation, e.g.,the poses of an object, mask images and 3D keypoints on the object. Using this pipeline, itis possible to adjust different settings to reduce the domain gap between synthetic data andreal-world data and get better pose estimation results. Such settings could be changing themethod of extracting 3D keypoints and varying the scale of the object or the light settingsin the scene.The network used to test the performance of training on our synthetic data is PVNet,which achieves state-of-the-art results for 6D pose estimation. This architecture learns tofind 2D keypoints of the object in the image, as well as 2D–3D keypoint correspondences.With these correspondences, the Perspective-n-Point (PnP) algorithm is used to extract apose. We evaluate the pose estimation of the different settings on the synthetic data andcompare these results to other state-of-the-art work. We find that using only real-worlddata for training is worse than using a combination of synthetic and real-world data. Sev-eral other findings are that varying scale and lightning, in addition to adding random back-ground images to the rendered images improves results. Four different novel keypoint se-lection methods are introduced in this work, and tried against methods used in previouswork. We observe that our methods achieve similar or better results. Finally, we use thebest possible settings from the synthetic data pipeline, but with memory limitations on theamount of training data. We are close to state-of-the-art results, and could get closer withmore data.

No description available.