Освітня програма наук про дані
https://er.ucu.edu.ua:443/handle/1/1187
Data Science Education Program2024-03-29T10:34:39ZWeakly-supervised tumor segmentation in computed tomography scans
https://er.ucu.edu.ua:443/handle/1/3951
Weakly-supervised tumor segmentation in computed tomography scans
Zakharchenko, Iryna
This research focuses on the topic of weakly-supervised tumor segmentation. The
proposed pipeline involves the usage of a classification model to make predictions
regarding the presence of a tumor in an image. Subsequently, the CAM (Class Activation Mapping) approach is employed to identify the most relevant regions within
the image as determined by the model. The underlying concept is that the model will
learn to identify tumor regions, resulting in higher activations in those areas. The
advantage of the weakly supervised approach is its ability to learn from a smaller
dataset, requiring only image-level labels in our case. By implementing the proposed pipeline, specifically using the Score-CAM technique.
2023-01-01T00:00:00ZMonitoring Sea Water Pollution from Satellite data
https://er.ucu.edu.ua:443/handle/1/3950
Monitoring Sea Water Pollution from Satellite data
Stepanov, Yevhen
This study is motivated by the Ocean Decade declared by the United Nations. We
employed machine learning techniques to detect and delineate areas of pollution in
the coastal zone of Great Britain, utilizing pollution reports from the Department for
Environment Food And Rural Affairs (DEFRA) and the ocean monitoring datasets
from the European Space Agency (ESA).
In this study, feature engineering was performed on chlorophyll concentration
data. Two datasets were constructed: one with statistical metrics (mean, median,
standard deviation, and percentiles) as features, and another with individual cells
of the chlorophyll concentration matrix as features, utilizing different matrix n sizes,
where n is from 3 to 11, where each element or pixel of the matrix represented a
1km × 1km area. Logistic regression, decision trees, random forest classifier, gradient boosting classifier, and LeNet models were applied. Hyper parameter tuning
was conducted to optimize the performance of each model. Among the models,
the gradient boosting classifier achieved the highest accuracy of 95.21%. Additionally, the F1 score was determined to be 0.2445, the ROC AUC was 0.7659, and the
precision-recall AUC (PR-AUC) was found to be 0.1821.
Detecting and delineating areas of pollution can greatly assist cleaning services
in efficiently carrying out their job, resulting in improved remediation and restoration efforts. The identification of pollution areas holds significant implications for
the fishing industry, as it enables informed decision-making regarding fishing practices and resource management, ensuring the sustainability and viability of the sector. Moreover, the accurate detection and delineation of pollution areas have the potential to generate substantial economic, social, and environmental benefits by facilitating targeted interventions, protecting ecosystems, preserving marine resources,
and fostering a healthier and more resilient environment.
The findings of this study provide valuable insights into the efficacy of classification approaches in identifying and mapping pollution sites in coastal regions using
pollution reports from DEFRA.
2023-01-01T00:00:00ZCorner localization and camera calibration from imaged lattices
https://er.ucu.edu.ua:443/handle/1/3949
Corner localization and camera calibration from imaged lattices
Stadnik, Andrii
This thesis proposes a model-based approach to improve the detection of calibration board fiducials from calibration imagery taken by wide-angle or fisheye lenses.
From a single image, we estimate the camera model and project the calibration board
into the image to guide the search for missed detections and reject spurious detections. In addition, we propose a classifier to label ambiguous detections that are
geometrically plausible given the estimated camera model and imaged board. The
proposed method addresses shortcomings of the state-of-the-art, which struggle to
reliably detect board fiducials at the extents of the image, where the lens distortion
is most observable. The proposed method recovers additional corners that can be
used to place additional constraints on the non-convex camera calibration problem,
which improves the likelihood of convergence to a global minimum.
The code for this paper is available on GitHub.
2023-01-01T00:00:00ZOcean surface visibility prediction
https://er.ucu.edu.ua:443/handle/1/3948
Ocean surface visibility prediction
Prypeshniuk, Volodymyr
Seawater transparency is an indispensable ecological parameter with substantial impacts on the health and productivity of aquatic ecosystems. Its significance spans
across various industries, including environment protection, fishing and tourism.
The fluctuating nature of aquatic systems and their intricate interplay with human
activities often induce substantial variability in seawater transparency. This underlines the pressing necessity for effective predictive tools in the stewardship and
preservation of our invaluable water resources.
Despite the clear importance of water transparency, ocean forecasting remains
a considerably understudied field, some work has been done on using satellite for
monitoring, but literature is scarce for forecasting with only few simple models explored. There is an evident gap in research and tools focused on predicting changes
in this crucial ecosystem, underlining the novelty and urgency of our work.
In this research, we aim is to develop a forecasting model that not only excels
in precision and speed, but is also flexible enough to encompass a vast array of
potential future scenarios. We primarily employed SimVP, a spatio-temporal convolutional neural network, for ocean forecasting purposes. This model was trained
using the earth observation data from the Copernicus Marine Service. This data
were collected for 20 years of daily observation of water transparency in the marine
environment surrounding the UK, with a spatial resolution of 4km x 4km.
Our findings showed that SimVP substantially outperformed the baseline models (AutoRegressive Integrated Moving Average (ARIMA) and Simple Exponential
Smoothing (SES)) in predicting the next day seawater transparency, demonstrating
an improvement of 17.4%, and a notable reduction in the Root Mean Square Error
(RMSE) from 2.63 to 2.24, and improvement in inference time efficiency in 66.3 times
(334.6 -> 5.04 seconds). We show that this method better performs better on regions
with minor variation like Irish Sea or English Channel, and performs worse on regions with high variations like Atlantic Ocean or North Sea.
Our study demonstrates the advantage of adopting the spatio-temporal neural
network architectures for ocean monitoring and paves the way for future research
in adopting advanced machine learning techniques in this field.
2023-01-01T00:00:00ZAdvancing medical image segmentation via pseudo-labeling of public datasets
https://er.ucu.edu.ua:443/handle/1/3947
Advancing medical image segmentation via pseudo-labeling of public datasets
Mishchenko, Roman
Our study explores the difficulties and possible resolutions in the domain of medical
image segmentation, with a special emphasis on utilizing unlabeled public datasets
to improve tumor segmentation. We suggest a strategy that incorporates pseudolabeling methodologies with real-world data to enhance the learning potential of
segmentation models. Yet, the findings imply that while improvements in model
performance exist, they are not substantial. The research underscores the paramount
importance of data quality over quantity, emphasizing that image characteristics influence the effectiveness of the process more than the total number of images.
2023-01-01T00:00:00ZAstronomical Data Features Extraction and Citation Prediction
https://er.ucu.edu.ua:443/handle/1/3946
Astronomical Data Features Extraction and Citation Prediction
Kutsuruk, Vladyslav
Natural Language Processing methods present promising opportunities for analyzing astronomical data, enabling the extraction of essential information from vast
amounts of observations. Yet, applying these techniques to astronomical data presents
notable challenges, including the difficulty of astronomical terminology and the diverse range of data sources. In this research, we leverage multiple Natural Language
Processing techniques to extract information from astronomical observations with a
specific focus on predicting the future citation rate of astronomical telegrams. To
achieve this, we create a comprehensive dataset gathering astronomical messages
from various sources and utilize techniques such as Named Entity Recognition,
doc2vec, word2vec, and topic extraction. Along with this, we enhance the extracted
information by incorporating manually created features that capture the characteristics of astronomical telegrams beyond their direct context. These features aim to
provide a comprehensive representation of the messages. We then use all the extracted information to predict the future impact of the telegrams, as indicated by
their citation counts, using multiple Machine Learning techniques.
2023-01-01T00:00:00ZText-Guided 3D Synthesis with Latent Diffusion Models
https://er.ucu.edu.ua:443/handle/1/3945
Text-Guided 3D Synthesis with Latent Diffusion Models
Kovalenko, Danylo
The emergence of diffusion models has greatly impacted the field of deep generative
models, establishing them as a powerful family of models with unparalleled performance in various applications such as text-to-image, image-to-image, and text-toaudio tasks. In this work, we aim to propose a solution for text-guided 3D synthesis
using denoising diffusion probabilistic models, while minimizing the memory and
computational requirements. Our goal is to achieve high-quality and high-fidelity
3D object generation conditioned by text or a label in a number of seconds. We propose to use a triplane space parametrization in combination with a Latent Diffusion
Model (LDM) to generate smooth and coherent geometry. The LDM is trained on
the large-scale text-3d dataset and is used as a latent triplane texture generator. By
using a triplane space parametrization, we aim to improve the efficiency of the space
representation and reduce the computational cost of synthesis. We will also give a
theoretical justification that this kind of parametrization of 3d space is capable of
containing not only information about the geometry but also about the color and
reflectivity of the figure. Additionally, we use an implicit neural renderer to decode
geometry details from triplane textures.
2023-01-01T00:00:00ZText generation with control conditions compliance
https://er.ucu.edu.ua:443/handle/1/3944
Text generation with control conditions compliance
Konopatska, Oleksandra
Controllable text generation has emerged as a significant research area, allowing the
production of text with desired characteristics. In this work, we investigate the controllability of text generation, exploring the challenges of controlling various aspects
of generated text, such as length, parts-of-speech (POS) structure, sentiment, and
tense; in addition, we extend our analysis to the task of multi-conditional text generation, which entails the possibility of simultaneous control of several parameters
of the generated text.
Our research is mainly based on fine-tuned GPT-2, an autoregressive transformerbased model. Using fine-tuned GPT-2, we managed to achieve notable progress in
controlling the above-mentioned text attributes; we also present the results of experiments using other approaches, such as diffusion models and ChatGPT. The models
are trained on our own dataset, meticulously curated in-house; the evaluation of
the generation results is carried out using a comprehensive set of control, fluency,
distinctiveness, and repetition metrics.
Through rigorous analysis, we assess the performance of studied models in terms
of controllability. Length control, in particular, proved to be a challenging aspect,
even when employing the largest available models. Nevertheless, our fine-tuned
GPT-2 demonstrated promising results, showcasing its capabilities in generating text
with desired characteristics.
Overall, our findings highlight the possibilities of controllable text generation
using fine-tuned GPT-2 and other models. Our work contributes to the ongoing exploration of techniques for improving controllability in text generation. As this field
continues to evolve, further research can build upon our analysis and methodologies
to enhance controllability and pave the way for more sophisticated text generation
systems.
2023-01-01T00:00:00ZSplit Activation Networks for Neural Fields
https://er.ucu.edu.ua:443/handle/1/3943
Split Activation Networks for Neural Fields
Kilianovskyi, Mykhailo
Neural field modeling is a developing area that improves state-of-the-art results in
tasks such as 3D scene reconstruction, image manipulation, generative modeling,
and other aspects of deep learning. In this work, we present SplitNet, a novel neural network architecture for neural field modeling that combines multiple activation
functions in a single layer. We try different techniques to improve performance,
such as proper weight initialization, and benchmark its performance on image representation, 3D scene reconstruction, and image classification tasks. As a part of the
work, we found a way to improve the performance of previous work on implicit
neural networks with sinusoidal activations in a limited setting and study how well
this improvement generalizes to other tasks and data.
2023-01-01T00:00:00ZAudio spoofing detection
https://er.ucu.edu.ua:443/handle/1/3942
Audio spoofing detection
Ivashchenko, Dmytro
Efficient and accurate audio spoofing detection is crucial to ensuring audio-based
systems’ security and integrity. Existing methods often mainly focused on the performance of the detection system. This master thesis focuses on the development of
advanced techniques that prioritize efficiency while maintaining high detection performance. We introduced the model, consisting of an encoder and a classifier, which
can efficiently learn complex representations with a lack of labeled data. We introduce suitable loss functions to effectively distinguish spoofed and bonafide speech
in latent space to keep the performance high. The results demonstrate notable improvements in both encoder performance and classification accuracy, highlighting
the potential for enhanced self-supervised audio analysis techniques.
Keywords: self-supervised learning, audio spoofing detection, automatic speaker
verification, audio processing, speech classification
2023-01-01T00:00:00ZThe 4th Stage of Genocide: Computational Analysis of Dehumanization of Ukrainians on Russian Telegram
https://er.ucu.edu.ua:443/handle/1/3941
The 4th Stage of Genocide: Computational Analysis of Dehumanization of Ukrainians on Russian Telegram
Burovova, Kateryna
Dehumanization is a pernicious psychological process of denying some or all attributes of humanness to the target group. It is frequently cited as a common hallmark of incitement to commit genocide. The international security landscape has
seen a dramatic shift following the 2022 Russian invasion of Ukraine. This, coupled
with recent developments in the conceptualization of dehumanization, necessitates
the creation of new techniques. These techniques need to be capable of analyzing and detecting this extreme violence-related phenomenon on a large scale. Our
project aims to pioneer the development of a detection system for instances of dehumanization in the Russian language. To achieve this, we collected the entire posting
history of the most popular political bloggers on Russian Telegram. We tested classical machine learning, deep leaning, and zero-shot learning approaches and applied
semantic modeling to explore the evolution of dehumanizing rhetoric. We found
that transformer-based method for entity extraction shows promising results for binary dehumanization classification when applied via an indicator mapping function, while additionally allowing for evaluation of the type of dehumanization instance. The proposed methods can be built into systems of anticipatory governance,
contribute to the collection of evidence of genocidal intent in the Russian invasion
of Ukraine, and pave the way to the large-scale studies of dehumanizing language
and representation of Ukrainians in Russian media.
2023-01-01T00:00:00ZHidden state refinement for optical flow forecasting
https://er.ucu.edu.ua:443/handle/1/3930
Hidden state refinement for optical flow forecasting
Babenko, Anton
In recent years the topic of optical flow has become well-spread due to computation
power support and optical flow estimation applications used on mobile phones and
edge devices: video editors, frame stabilizations, and autonomous driving feature
providers. This work analyzes multiple approaches to optical flow estimation and
finds the main problems of the optical flow methods: slow convergence and long
execution of the prediction algorithm. We propose to solve the slow convergence
and long execution time with hidden state refinement to provide the initialization
for optical flow estimation based on several previous frames and their hidden state
transformations, which imitates the pixel movement at the hidden state level. The
proposed method uses CNN, LSTM, and Transformer blocks which help to achieve
the optical flow estimation and hidden state refinement to speed up the system. We
used Sintel, KITTY-15, FlyingChairs, FlyingThings, HD1K, DAVIS, and YouTubeVOS datasets for our experiment.
2023-01-01T00:00:00ZLanguage-Agnostic detection of Current Events across Wikipedia
https://er.ucu.edu.ua:443/handle/1/3922
Language-Agnostic detection of Current Events across Wikipedia
Antypova, Alisa
Currently, English Wikipedia alone includes over 6,660,000 articles and it averages
550 new articles per day. To assist the readers in identifying pages that cover recent
noteworthy occurrences the Current Events portal was implemented. However, this
portal is maintained manually with notable quality differences across languages.
The main goal of this work is to establish the task of supervised event detection in
Wikipedia and propose a language-agnostic solution to address this problem. This
is an important milestone towards improving the quality of the Current Events Portal
for the languages with not many active editor communities. In this work, we
reviewed existing research on this topic, and by combining and enriching those existing
solutions, we proposed a current event detection dataset based on the Current
Events Portal updates and Wikipedia pages’ features. Also, we developed a
language-agnostic event detection model and reported its performance in English,
German, and Polish languages, showing that is possible to automatize this task. The
outcome of this work can be used to assist Wikipedia editors to keep the Current
Event portal updated, saving time and the human effort used on this task.
2023-01-01T00:00:00ZOne-shot Facial Expression Reenactment using 3D Morphable Models
https://er.ucu.edu.ua:443/handle/1/3168
One-shot Facial Expression Reenactment using 3D Morphable Models
Vei, Roman
The recent advance in generative adversarial networks has shown promising results
in solving the problem of head reenactment. It aims to generate novel images with
altered poses and emotions while preserving the identity of a human head from a
single photo. Current approaches have limitations, making them inapplicable for
real-world applications. Specifically, most algorithms are computationally expensive,
have no apparent tools for manual image manipulation, require audio or take
multiple input images to generate novel images.
Our method addresses the single-shot face reenactment problem with an end-toend
algorithm. The proposed method utilizes head 3D morphable model (3DMM)
parameters to encode identity, pose, and expression. With the proposed approach,
the pose and emotion of a person on an image is changed by manipulating its 3DMM
parameters. Our work consists of a face mesh prediction network and a GAN-based
renderer. A predictor is a neural network with simple encoder architecture that regresses
3D mesh parameters. A renderer is a GAN network with warping and rendering
submodules that renders images from a single source image and target image
3DMM parameters.
This work proposes a novel head reenactment framework that is computationally
efficient and uses 3DMM parameters that are easy to alter, making the proposed
method applicable in real-life applications. It is first to our knowledge approach
that simultaneously solves two of these problems: 3DMM parameters prediction
and face reenactment, and benefits from both.
2022-01-01T00:00:00ZGeneralizing texture transformers for super-resolution and inpainting
https://er.ucu.edu.ua:443/handle/1/3167
Generalizing texture transformers for super-resolution and inpainting
Romanus, Teodor
The new multi-camera smartphones and recent advancements in generalized
Machine Learning models make it possible to bring new types of photo editing neural
networks to the market. This thesis covers methods of image enhancement with
texture transfer. The known high-resolution regions (reference) can be utilized to
restore degraded areas of an image. The task of restoring partially degraded images
can be defined “partial super-resolution.” The task of restoring missing parts of
images is called inpainting. We propose to use the novel Texture Transformer Network
for Image Super-Resolution (TTSR) to solve the partial super-resolution and
inpainting tasks.
The fully convolutional networks are unable to copy image patches. This inability
forces the model to store textures using the train weights. The usage of the attention
mechanism allows taking advantage of joint feature learning in low-resolution
and high-resolution parts of images simultaneously, in which deep feature correspondences
can be discovered by attention. This approach exhibits an accurate
transfer of texture features.
The experiments confirm that the TTSR network can be used to solve the partial
super-resolution and inpainting tasks simultaneously. Modifications of the network
(different embedding sizes, soft-attention, trainable projections) study the architecture
capacity to solve the specified tasks. The evaluation of results includes comparing
the TTSR network with an inpainting network for the inpainting task.
2022-01-01T00:00:00ZMulti-temporal Satellite Imagery Panoptic Segmentation of Agricultural Land in Ukraine
https://er.ucu.edu.ua:443/handle/1/3166
Multi-temporal Satellite Imagery Panoptic Segmentation of Agricultural Land in Ukraine
Petruk, Marian
Remote sensing of the Earth using satellites helps analyze the Earth’s resources,
monitor local land surface changes, and study global climate changes. In particular,
farmland information helps farmers in decision-making, planning and increases
productivity to achieve better agro-ecological conditions. In this work, we primarily
focus on panoptic segmentation of agricultural land, a combination of two parts:
1) delineation of parcels (instance segmentation) and 2) classification of parcel crop
type (semantic segmentation). Second, we explore how multi-temporal satellite imagery
data compares to a single image query in segmentation performance. Third,
we conduct experiments using the recent advances in Deep Learning and Computer
Vision that improve the performance of such systems. Finally, we show the performance
of the state-of-the-art panoptic segmentation algorithm on the agricultural
land of Ukraine, where the farmland market has just opened.
2022-01-01T00:00:00ZReinforcement Learning Agents in Procedurally-generated Environments with Sparse Rewards
https://er.ucu.edu.ua:443/handle/1/3165
Reinforcement Learning Agents in Procedurally-generated Environments with Sparse Rewards
Nahirnyi, Oleksii
Solving sparse-reward environments is one of the most considerable challenges for
state-of-the-art (SOTA) Reinforcement Learning (RL). Recent usage of sparse-rewards
in procedurally-generated environments (PGE) to more adequately measure agent’s
generalization capabilities via randomization makes this challenge even harder. Despite
some progress of newly created exploration-based algorithms in MiniGrid PGEs,
the task remains open for research in terms of improving sample complexity. We
contribute to solving this task by creating a new formulation of exploratory intrinsic
reward. We base this formulation on a thorough review and categorization of other
methods in this area. Agent that optimizes an RL objective with such a formulation
performs better than SOTA methods in some small or medium sized PGEs.
2022-01-01T00:00:00ZPolyp detection and segmentation from endoscopy images
https://er.ucu.edu.ua:443/handle/1/3158
Polyp detection and segmentation from endoscopy images
Kokshaikyna, Mariia
Endoscopy is a widely used clinical procedure for the detection of different diseases
in internal gastrointestinal tract’s organs such as the stomach and colon. Modern endoscopes
allow getting high-quality video during the procedure. Computer-assisted
methods might support medical specialists in detecting or segmenting anomaly regions
on the picture. Many datasets are available and methods to detect polyp regions
have been proposed. One kind of task is polyps segmentation on images and
videos. The best results in semantic segmentation of polyps are now achieved with
fully supervised approaches. In this thesis, we describe experiments with CaraNet
model. We checked robustness on cross-validation on several publicly available
datasets and small private dataset, tried a few modifications of attention layer in
order to improve performance, presented and discussed results.
2022-01-01T00:00:00ZBrain age prediction based on EEG records
https://er.ucu.edu.ua:443/handle/1/3157
Brain age prediction based on EEG records
Klymenko, Mykola
Ambulatory EEG is a widespread test used in hospitals for the neurological evaluation
of patients. EEG waveforms are typically reviewed by a trained neurologist
to classify EEG into clinical categories. Methodologically, there is a need to classify
EEG recordings automatically. Ideally, the classification models should be interpretable,
able to deal with EEG of varying durations, and robust to various artifacts.
We aimed to test and validate a framework for EEG classification, which satisfies
such requirements by symbolizing EEG signals and adapting a method previously
proposed in natural language processing (NLP).We considered an extensive sample
of routine clinical EEG (n=5’850), with a wide range of ages between 0 and 100 years
old. We symbolized the multi-variate EEG times series and applied a byte-pair encoding
(BPE) algorithm to extract a dictionary of the most frequent patterns (tokens)
reflecting the variability of EEG waveforms. To demonstrate the performance of
such an approach, we used newly-reconstructed EEG features to predict the biological
age of patients with Random Forest. We also correlated the relative frequencies
of tokens with age. We found that the age prediction model achieved the mean absolute
error of 15.9 in years. The correlation between actual and predicted age was
0.56. The most significant correlations between the frequencies of tokens and age
were observed at frontal and occipital EEG channels. Our findings demonstrate the
feasibility of an approach based on applying NLP methods to time series classification.
Notably, the proposed algorithms could be instrumental in classifying clinical
EEG with minimal preprocessing and sensitivity to the appearance of short events,
such as epileptic spikes.
2022-01-01T00:00:00ZObject detection in automotive vehicle domain based on real and synthetic data
https://er.ucu.edu.ua:443/handle/1/3156
Object detection in automotive vehicle domain based on real and synthetic data
Ilechko, Roman
In recent years, we have seen an incredible increase in deep learning. With increasing
interest, we also have an increasing number of bold and even revolutionary studies
that drive progress and boost models performance. Despite the fact that the number
of large and high-quality data-sets is growing rapidly, we could often observe that
models need even more data for many domains and tasks. Usually, additional data
is needed not only for giant models. Even though the domains like autonomous
vehicles, which mainly focus on lightweight models, require extra data. We should
state that sometimes data labeling is not a panacea. Especially for autonomous vehicles,
as the data provided must have a great variety and low error risk. The additional
synthetic could be an excellent booster for existing approaches or even a
must-have part of training data. For example, simulators give the ability to manage
the scene’s complexity by controlling the number of objects, their size, and their interaction
with the environment, which could be very helpful for such tasks as object
detection. Nowadays, the researcher should intuitively balance the ratio of natural
and generated data simultaneously, considering the possibility of gaps between the
two domains. Despite the fact that the mentioned task is not evident, constraints like
model size and count of classes could bring additional unclarity. In this paper, we
precisely analyze the impact of synthetic data on the training process, cover possible
training strategies, and provide guidance on defining the amount of artificial data
with existing constraints.
2022-01-01T00:00:00ZLarge-scale product classification for efficient matching in procurement systems
https://er.ucu.edu.ua:443/handle/1/3154
Large-scale product classification for efficient matching in procurement systems
Hrysha, Ihor
We consider the problem of recommending relevant suppliers given detailed request
context in a procurement setting. The fundamental recommendation in procurement
systems is that a single query has potentially hundreds of relevant suppliers associated.
A complicating factor is that, for most suppliers, we do not have a complete
listing of product and service offerings, in contrast with most literature in the space
of product search. An additional difficulty is introduced by the fact that queries are
generated by users operating within large procurement organizations, each building
queries in idiosyncratic but internally consistent ways, and each organizing activities
according to a unique internal product taxonomy. The central research question
that we aim to address is: can we utilize this vast but inconsistently structured set
of product data that allows us to derive semantic meaning across users and contexts?
We propose several fully and semi-supervised approaches and benchmark
them using a proprietary dataset that includes large-scale procurement data as well
as supplier-provided catalogs. Finally, and uniquely, we experimentally validate the
performance of our preferred model in a live production setting.
2022-01-01T00:00:00ZMobile Object Tracking with Siamese Neural Network
https://er.ucu.edu.ua:443/handle/1/3153
Mobile Object Tracking with Siamese Neural Network
Borsuk, Vasyl
Visual object tracking is one of the most fundamental research topics in computer
vision that aims to obtain the target object’s location in a video sequence given the
object’s initial state in the first video frame. The recent advance of deep neural networks,
specifically Siamese networks, has led to significant progress in visual object
tracking. Despite being accurate and achieving high results on academic benchmarks,
current state-of-the-art approaches are compute-intensive and have a large
memory footprint that cannot satisfy the strict performance requirements of realworld
applications. This work focuses on designing a novel lightweight framework
for resource-efficient and accurate visual object tracking. Additionally, we introduce
a new tracker efficiency benchmark and protocol where efficiency is defined in terms
of both energy consumption and execution speed on edge devices.
2022-01-01T00:00:00Z3D Head Model Estimation from a Single Photo
https://er.ucu.edu.ua:443/handle/1/2710
3D Head Model Estimation from a Single Photo
Zatserkovnyi, Rostyslav
Today, 3D human head models are widely used in fields such as computer vision, entertainment,
healthcare, and biometrics. Since a high-quality scan of a human head
is expensive and time-consuming to obtain, machine learning algorithms are used
to estimate the shape and texture of a 3D model from a single "in-the-wild" photograph,
often taken at extreme angles or with non-uniform illumination. However,
as a full head texture cannot be trivially inferred from a single photograph due to
self-occlusion, many only focus on modeling an incomplete and partially textured
model of the human head.
This work proposes a machine learning pipeline that reconstructs a fully textured
3D head model from a single photograph. We collect a novel dataset of 99.3
thousand high-resolution human head textures created from synthetic celebrity photographs.
To the best of our knowledge, this is the first UV texture dataset of a similar
scale and fidelity. Using this dataset, we train a free-form inpainting GAN that
learns to recreate full head textures from partially obscured projections of the input
photograph.
2021-01-01T00:00:00ZReal-time inverse kinematics and inverse dynamics from motion capture
https://er.ucu.edu.ua:443/handle/1/2709
Real-time inverse kinematics and inverse dynamics from motion capture
Zabava, Kateryna
This work applies machine learning to solving inverse dynamics and inverse kinematics
tasks from the motion capture data. This approach may simplify the calculation
process and help do scientific simulations as part of a physics engine that
describes the neural control of human motion and decodes movement intent in individuals
with neural damage. The existing algorithm has to be modified for every
experiment and takes a significant amount of time to execute. It is also sensitive to
noise and missing data, and it is not a real-time calculation. We propose a solution
of inverse kinematics tasks with neural networks. Here we report accuracy results
both on clean data and noisy data. We also apply a similar approach for the inverse
dynamics task. The approach shows high accuracy on clean data, but this accuracy
decreases if applied to the noisy data.
2021-01-01T00:00:00ZAutomated Fact-checking for Wikipedia
https://er.ucu.edu.ua:443/handle/1/2708
Automated Fact-checking for Wikipedia
Trokhymovych, Mykola
The incoming flow of information is continuously increasing along with the disinformation
piece that can harm society. Filtering unreliable content helps keep
Wikipedia as free as possible of disinformation, making it one of the most significant
reliable information sources. Consequently, Wikipedia’s knowledge base is widely
used for facts verification academic research. The main goal of our work is to transform
recent academic achievements into a practical open-source Wikipedia-based
fact-checking application that is both accurate and efficient. We review the primary
NLI related datasets and study their relevant limitations. As a result, we propose the
data filtering method that improves the model’s performance and generalization.
We show that transfer learning for NLI models are not working well, and complete
model training is needed to achieve the best result on a specific dataset. We come up
with an unsupervised fine-tuning of the Masked Language model on field-specific
texts for model domain adaptation. Finally, we present the new fact-checking system
WikiCheck API that automatically performs a facts validation process based on the
Wikipedia knowledge base. It is comparable to SOTA solutions in terms of accuracy
and can be used on low memory CPU instances.
2021-01-01T00:00:00ZImproving Sequence Tagging for Grammatical Error Correction
https://er.ucu.edu.ua:443/handle/1/2707
Improving Sequence Tagging for Grammatical Error Correction
Tarnavskyi, Maksym
In this work, we investigated the recent sequence tagging approach for the Grammatical
Error Correction task. We compared the impact of different transformerbased
encoders of base and large configurations and showed the influence of tags’
vocabulary size. Also, we discovered ensembling methods on data and model levels.
We proposed two methods for selecting better quality data and filtering noisy
data. We generated new training GEC data based on knowledge distillation from an
ensemble of models and discovered strategies for its usage. Our best ensemble without
pre-training on the synthetic data achieves a new SOTA result of an F0.5 76.05 on
BEA-2019 (test), in contrast, when the newest obtained results were achieved with
pre-training on synthetic data. Our best single model with pre-training on synthetic
data achieves F0.5 of 73.21 on BEA-2019 (test). Our investigation improved the previous
results by 0.8/2.45 points for the single/ensemble sequence tagging models.
The code, generated datasets, and trained models are publicly available.
2021-01-01T00:00:00ZModeling and Prediction of Alzheimer’s Disease Progression
https://er.ucu.edu.ua:443/handle/1/2706
Modeling and Prediction of Alzheimer’s Disease Progression
Smailova, Sevil
Alzheimer’s Disease is an irreversible disease that causes a decline in cognitive
abilities and leads to dementia. Many efforts are applied to understand the behavior
of the disease progression and foresee its future state. The metrics that assess the
level of cognition are named as cognitive scores. The dynamics of cognitive scores
help understand the future disease progression. However, there is a lack of understanding
on what is the best benchmark for the predicted value of the cognitive
score. Moreover, there could be cases when the future value of the cognitive score is
not statistically different comparing to the current value.
In this work we discover those patients that by design cannot have the dynamics
in their progression of cognitive scores. We justify that the dynamics of progression
for Cognitively Normal patients do not change over five years. We reveal that there
is no statistically significant change in progression after the 1-year follow-ups. We
unified the evaluation framework of different imputation, feature selection methods
and machine learning models on different time to prediction settings as well as on
different patient populations.
2021-01-01T00:00:00ZCentral pattern generator model using spiking neural networks
https://er.ucu.edu.ua:443/handle/1/2705
Central pattern generator model using spiking neural networks
Pryyma, Yuriy
are essential for survival. The neural locomotor pathways contain the central
pattern generator (CPG), a network of neurons embedded into the spinal
cord and generating dynamic output for walking and running. Even though
there are multiple formulations of the CPG, from coupled oscillators to complex
networks of Hodgkin-Huxley neurons, the optimal choice of model implementation
depends on its application. The choice of a formulation is often
described as the trade-off between complexity and the level of details in
the model’s function. However, the advantages between different formulations
have not been established. Recently, the spiking neural networks (SNN)
have gained popularity as a biological analog for neural dynamics that uses
methodology developed for artificial neural networks. This formulation uses
spiking frequency instead of rate signals to accomplish dynamic computations
with the integrate-and-fire neurons.
In this study1, we aimed to create the framework for comparing a versatile
CPG rate model and its implementation with the model build with SNN.
We used a neuromorphic software package (Nengo) to develop and validate
a bilateral CPG model’s structural and functional details based on the halfcenter
oscillators. The spiking model shows similar precision for calculating
the empirical phase-duration characteristics of gait in cats as the rate model,
and it also reproduces the linear relationship between the CPG input and the
empirical limb speed of forward progression. While the phase characteristic
was used to optimize neural dynamics, the input relationship with the
limb speed is the product of the model structure. Furthermore, the spiking
model has increased tolerance to temporal noise, and it can withstand some
structural damage. The spiking and rate models require further comparative
analysis. Overall, the development of adaptable spiking models could
help integrate the biomimetic components within the control systems for assistive
robotics and electrical stimulation devices to rehabilitate locomotion
after central and peripheral injuries.
2021-01-01T00:00:00ZIncorporating Metadata for Semantic Segmentation by employing Channel Attention Mechanism
https://er.ucu.edu.ua:443/handle/1/2704
Incorporating Metadata for Semantic Segmentation by employing Channel Attention Mechanism
Plutenko, Iaroslav
The meta-information accompanying data from image acquisition devices has limited
use in microscopy image processing techniques involving Deep Learning. This
project aims to incorporate the supplementary metadata for semantic segmentation
by employing a channel selection mechanism in convolutional networks outlining
its potential benefits and practical applications where metadata can be used
for switching tasks within a master model. The results of conducted experiments
show that meta-information is helpful, and the phenomenon is more expressed with
incompatible segmentation tasks, where a multi-head model or separate models are
required otherwise. Overall, we have achieved a slight increase in scores for similar
tasks as well and demonstrated the applicability of the CNN model for separate
tasks, forcing it to work as an ensemble, leveraging the beneficial effect of multi-task
learning.
2021-01-01T00:00:00ZRobust Visual Odometry for Realistic PointGoal Navigation
https://er.ucu.edu.ua:443/handle/1/2703
Robust Visual Odometry for Realistic PointGoal Navigation
Partsey, Ruslan
The ability to navigate in complex environments is a fundamental skill of a home
robot. Despite extensive study, indoor navigation in unseen environments under
noisy actuation and sensing and without access to precise localization continues
to be an open frontier for research in Embodied AI. In this work, we focus on designing
a visual odometry module for robust egomotion estimation and it’s integration
with navigation policy for efficient navigation under noisy actuation and sensing.
Specifically, we study how the observations transformations and incorporating
meta-information available to the navigation agent impacts visual odometry model
generalization performance. We present a set of regularization techniques that can
be implemented as train- and test-time augmentations to increase the robustness to
noise.
Navigation agent, equipped with our visual odometry module, reaches the goal
in 86% of episodes and scores 0.66 SPL in Habitat Challenge 2021 benchmark.
2021-01-01T00:00:00ZPoint cloud human pose estimation using capsule networks
https://er.ucu.edu.ua:443/handle/1/2702
Point cloud human pose estimation using capsule networks
Onbysh, Oleksandr
Human pose estimation based on points cloud is an emerging field that develops
with 3D scanning devices’ popularity. Build-in LiDAR technology in mobile phones
and a growing VR market creates a demand for lightweight and accurate models
for 3D point cloud. Widely advanced deep learning tools are mainly used for structured
data, and they face new challenges in unstructured 3D space. Recent research
on capsule networks proves that this type of model outperforms classical CNN architectures
in tasks that require viewpoint invariance from the model. Thus capsule
networks challenge multiple issues of classic CNNs like preserving the orientation
and spatial relationship of extracted features, which could significantly improve the
3D points cloud classification task’s performance.
The project’s objective is to experimentally assess the applicability of capsule
neural network architecture to the task of point cloud human pose estimation and
measure performance on non-synthetic data. Additionally, measure noise sustainability
of capsule networks for 3D data compared to regular models. Compare models’
performance with restricted amount of training data.
2021-01-01T00:00:00ZInvestment modeling of agricultural land valuation in Ukraine
https://er.ucu.edu.ua:443/handle/1/2701
Investment modeling of agricultural land valuation in Ukraine
Novosad, Nataliia
2021-01-01T00:00:00ZReal-time simulation of arm and hand dynamics using ANN
https://er.ucu.edu.ua:443/handle/1/2700
Real-time simulation of arm and hand dynamics using ANN
Manukian, Mykhailo
The physics of body dynamics is a complex problem solved by the nervous system
in real-time during the planning and execution of movements. The human arm
and hand have complex mechanics involving hundreds of muscles that actuate over
30 degrees of freedom (DOF). To date, the problems of this complexity remain unsolved
in engineering; yet, the nervous system computes control signals in a robust,
accurate, and time-efficient manner. Neuroprosthetics require similar computations
for the decoding of intent and encoding of sensory feedback. The trade-off of required
computational accuracy and latency is hard to resolve with classical physics;
thus, this research aims to develop "good-enough" approximations of these computations
using machine learning methods, such as artificial neural networks (ANN).
The kinematic and kinetic temporal computations that rely on the diverse number
of terms within the equations of motion are consistent with the recurrent neural
network (RNN) architectures. This study will test the general hypothesis that the
inverse dynamics of arm and hand can be captured with RNN formulation and explore
the utility of different architectures: i) simple Recurrent ANN, ii) Gated Recurrent
Unit (GRU) ANN, and iii) Long Short-Term Memory (LSTM) ANN. The inverse
problem is the mapping from joint kinematics (position, velocity, acceleration)
to joint kinetics (torque). The training and testing datasets were derived from the
physical model of arm and hand performing point-to-point movements between realistic
postures arranged in a grid within the physiological range of motion. Lastly,
we assessed the execution latency of the machine learning solutions in the context
of real-time requirements for prosthetic applications.
2021-01-01T00:00:00ZBLE Mesh Reliability Optimization using Neural Networks
https://er.ucu.edu.ua:443/handle/1/2699
BLE Mesh Reliability Optimization using Neural Networks
Bratus, Oleksandr
The Bluetooth Low Energy (BLE) Mesh network technology is one of the newest
technologies in the wireless communication domain. Due to low cost and low power
consumption, it has already become widespread and has the potential for a wide
range of applications.
However, the flooding algorithm on which based BLE Mesh data transmission
process impacts strongly on networks reliability. Because improper network setup
can be critical to ensuring sufficient network reliability, it is necessary to be able
to predict the network reliability in order to be able to reconfigure the network to
improve its reliability.
In this master thesis, we propose neural network approaches that predict the
reliability of both the entire network and its individual nodes. Presented results
demonstrate that trained neural networks are scalable by providing high accuracy
of predictions on networks of different sizes.
2021-01-01T00:00:00ZDetecting patterns of coordinated news article dissemination
https://er.ucu.edu.ua:443/handle/1/2697
Detecting patterns of coordinated news article dissemination
Bodnar, Petro
2021-01-01T00:00:00ZRaspberry quality detection in visual spectrum using neural networks
https://er.ucu.edu.ua:443/handle/1/2696
Raspberry quality detection in visual spectrum using neural networks
Blagodyr, Andrii
2021-01-01T00:00:00ZPredicting Properties of Crystals
https://er.ucu.edu.ua:443/handle/1/2242
Predicting Properties of Crystals
Lapchevskyi, Kostiantyn
2020-01-01T00:00:00ZNeural architecture search: a probabilistic approach
https://er.ucu.edu.ua:443/handle/1/2241
Neural architecture search: a probabilistic approach
Lut, Volodymyr
2020-01-01T00:00:00ZEnhancing controllability of text generation
https://er.ucu.edu.ua:443/handle/1/2240
Enhancing controllability of text generation
Shcherbyna, Anton
2020-01-01T00:00:00ZContext Independent Speaker Classification
https://er.ucu.edu.ua:443/handle/1/2052
Context Independent Speaker Classification
Olshanetskyi, Borys
2020-01-01T00:00:00ZMatching Red Links with Wikidata Items
https://er.ucu.edu.ua:443/handle/1/2051
Matching Red Links with Wikidata Items
Liubonko, Kateryna
2020-01-01T00:00:00ZMeme Generation for Social Media Audience Engagement
https://er.ucu.edu.ua:443/handle/1/2050
Meme Generation for Social Media Audience Engagement
Kurochkin, Andrew
2020-01-01T00:00:00ZRegion-Selected Image Generation with Generative Adversarial Networks
https://er.ucu.edu.ua:443/handle/1/2049
Region-Selected Image Generation with Generative Adversarial Networks
Korshunov, Vadym
2020-01-01T00:00:00ZEfficient Generation of Complex Data Distributions
https://er.ucu.edu.ua:443/handle/1/2048
Efficient Generation of Complex Data Distributions
Kofman, Philipp
2020-01-01T00:00:00ZStatistical and neural language models for the Ukrainian Language
https://er.ucu.edu.ua:443/handle/1/2047
Statistical and neural language models for the Ukrainian Language
Khaburska, Anastasiia
2020-01-01T00:00:00ZCustomer Lifetime Value for Retail Based on Transactional and Loyalty Card Data
https://er.ucu.edu.ua:443/handle/1/2046
Customer Lifetime Value for Retail Based on Transactional and Loyalty Card Data
Kasprova, Anastasiia
2020-01-01T00:00:00ZUnsupervised text simplification using neural style transfer
https://er.ucu.edu.ua:443/handle/1/2045
Unsupervised text simplification using neural style transfer
Kariuk, Oleg
2020-01-01T00:00:00ZReplica Exchange For Multiple-Environment Reinforcement Learning
https://er.ucu.edu.ua:443/handle/1/2044
Replica Exchange For Multiple-Environment Reinforcement Learning
Glusco, Dmitri
2020-01-01T00:00:00ZConcept embedding and network analysis of scientific innovations emergence
https://er.ucu.edu.ua:443/handle/1/2043
Concept embedding and network analysis of scientific innovations emergence
Brodiuk, Serhii
2020-01-01T00:00:00ZDetermining sentiment and important properties of Ukrainian-language user reviews
https://er.ucu.edu.ua:443/handle/1/2042
Determining sentiment and important properties of Ukrainian-language user reviews
Babenko, Dmytro
2020-01-01T00:00:00ZMinimal Solvers for Single-View Auto-Calibration
https://er.ucu.edu.ua:443/handle/1/2039
Minimal Solvers for Single-View Auto-Calibration
Lochman, Yaroslava
2020-01-01T00:00:00ZReinforcement Learning for Voltage Control-based Ancillary Service using Thermostatically Controlled Loads
https://er.ucu.edu.ua:443/handle/1/2038
Reinforcement Learning for Voltage Control-based Ancillary Service using Thermostatically Controlled Loads
Lukianykhin, Oleh
2020-01-01T00:00:00ZEnsembling and transfer learning for multi-domain microscopy image segmentation
https://er.ucu.edu.ua:443/handle/1/2037
Ensembling and transfer learning for multi-domain microscopy image segmentation
Misko, Oleh
2020-01-01T00:00:00ZStock market prediction utilizing central bank’s policy statements
https://er.ucu.edu.ua:443/handle/1/2036
Stock market prediction utilizing central bank’s policy statements
Moiseiev, Roman
2020-01-01T00:00:00ZImage Recommendation for Wikipedia Articles
https://er.ucu.edu.ua:443/handle/1/1920
Image Recommendation for Wikipedia Articles
Onyshchak, Oleh
2020-01-01T00:00:00ZTopological approach to Wikipedia article recommendation
https://er.ucu.edu.ua:443/handle/1/1917
Topological approach to Wikipedia article recommendation
Opirskyi, Maksym
2020-01-01T00:00:00ZAudience profile construction for local businesses marketing campaigns
https://er.ucu.edu.ua:443/handle/1/1916
Audience profile construction for local businesses marketing campaigns
Ovchynnikov, Kostiantyn
2020-01-01T00:00:00ZParameterizing Human Speech Generation
https://er.ucu.edu.ua:443/handle/1/1908
Parameterizing Human Speech Generation
Perepichka, Nazariy
2020-01-01T00:00:00ZGeneration of sport news articles from match text commentary
https://er.ucu.edu.ua:443/handle/1/1905
Generation of sport news articles from match text commentary
Porplenko, Denys
2020-01-01T00:00:00ZChanging clothing on people images using generative adversarial networks
https://er.ucu.edu.ua:443/handle/1/1904
Changing clothing on people images using generative adversarial networks
2020-01-01T00:00:00ZPerson re-identification in a top-view multi-camera environment
https://er.ucu.edu.ua:443/handle/1/1903
Person re-identification in a top-view multi-camera environment
Prodaiko, Ivan
2020-01-01T00:00:00Z3D Reconstruction of Video Sign Language Dictionaries
https://er.ucu.edu.ua:443/handle/1/1902
3D Reconstruction of Video Sign Language Dictionaries
Riazantsev, Roman
2020-01-01T00:00:00ZA multifactorial optimization of personnel scheduling in fleets of seagoing vessels
https://er.ucu.edu.ua:443/handle/1/1899
A multifactorial optimization of personnel scheduling in fleets of seagoing vessels
Smyrnov, Oleksandr
2020-01-01T00:00:00ZContext-Based Question-Answering System for the Ukrainian Language
https://er.ucu.edu.ua:443/handle/1/1898
Context-Based Question-Answering System for the Ukrainian Language
Tiutiunnyk, Serhii
2020-01-01T00:00:00ZBuilding segment based revenue prediction for CLV model
https://er.ucu.edu.ua:443/handle/1/1339
Building segment based revenue prediction for CLV model
Zorina, Kateryna
2019-01-01T00:00:00ZAspects of software naturalness through the generation of identifier names
https://er.ucu.edu.ua:443/handle/1/1338
Aspects of software naturalness through the generation of identifier names
Zaitsev, Oleksandr
2019-01-01T00:00:00ZMusic Generation Powered by Artificial Intelligence
https://er.ucu.edu.ua:443/handle/1/1337
Music Generation Powered by Artificial Intelligence
Shyshkin, Oleh
2019-01-01T00:00:00ZDetection of Difficult for Understanding Medical Words using Deep Learning
https://er.ucu.edu.ua:443/handle/1/1336
Detection of Difficult for Understanding Medical Words using Deep Learning
Pylieva, Hanna
2019-01-01T00:00:00ZSemi-supervised feature sharing for efficient video segmentation
https://er.ucu.edu.ua:443/handle/1/1335
Semi-supervised feature sharing for efficient video segmentation
Ponomarchuk, Anton
2019-01-01T00:00:00ZAutomatic Plant Counting using Deep Neural Networks
https://er.ucu.edu.ua:443/handle/1/1334
Automatic Plant Counting using Deep Neural Networks
Pidhirniak, Oleh
2019-01-01T00:00:00ZConvolutional Graph Embeddings for article recommendation in Wikipedia
https://er.ucu.edu.ua:443/handle/1/1333
Convolutional Graph Embeddings for article recommendation in Wikipedia
Moskalenko, Oleksii
2019-01-01T00:00:00ZMulti-task learning for image restoration
https://er.ucu.edu.ua:443/handle/1/1332
Multi-task learning for image restoration
Martyniuk, Tetiana
2019-01-01T00:00:00ZColor and style transfer using Generative Adversarial Networks
https://er.ucu.edu.ua:443/handle/1/1331
Color and style transfer using Generative Adversarial Networks
Kusyy, Andriy
2019-01-01T00:00:00ZCustomer Lifetime Value For Credit Limit Optimization
https://er.ucu.edu.ua:443/handle/1/1330
Customer Lifetime Value For Credit Limit Optimization
Kostiv, Markiyan
2019-01-01T00:00:00ZSemantic segmentation for visual indoor localization
https://er.ucu.edu.ua:443/handle/1/1329
Semantic segmentation for visual indoor localization
Kaminskyi, Yurii
2019-01-01T00:00:00ZStable and efficient video segmentation via GAN predicting adjacent frame
https://er.ucu.edu.ua:443/handle/1/1328
Stable and efficient video segmentation via GAN predicting adjacent frame
Ilnytskyi, Ivan
2019-01-01T00:00:00Z3D Hand Pose Estimation from Single RGB Camera
https://er.ucu.edu.ua:443/handle/1/1327
3D Hand Pose Estimation from Single RGB Camera
Chernytska, Olha
2019-01-01T00:00:00ZSafe Augmentation: Learning Task-Specific Transformations from Data
https://er.ucu.edu.ua:443/handle/1/1312
Safe Augmentation: Learning Task-Specific Transformations from Data
Baran, Irynei
2019-01-01T00:00:00ZGeneration of code from text description with syntactic parsing and Tree2Tree model
https://er.ucu.edu.ua:443/handle/1/1191
Generation of code from text description with syntactic parsing and Tree2Tree model
Stehnii, Anatolii
Software development requires vast knowledge of different programming tools which cannot be kept in human memory. Therefore software developers often formulate their task in human language to query online knowledge bases like StackOverflow to get short snippets of code. In this work, we explored the way of code generation from natural language description and prepared web API for Python which translates NL descriptions to short snippets of code. Our model implements sequence-to-sequence model with recursive encoder and uses syntactic trees instead of plain sequence on input. Results have not outperformed current state-of-the-art performance. However, presented Tree2Tree model has potential in other applications and this work makes a solid base for a further research.
2018-01-01T00:00:00ZApplication of Generative Neural Models for Style Transfer Learning in Fashion
https://er.ucu.edu.ua:443/handle/1/1190
Application of Generative Neural Models for Style Transfer Learning in Fashion
Mykhailych, Mykola
The purpose of this thesis is to analyze different generative adversarial networks for application in fashion. Research of “mode collapse” problem of generative adversarial networks. We studied the theoretical part of the “mode collapse” and conducted experiments on a synthetic toy dataset, and a dataset containing real data from fashion. Due to the developed method, it was possible to achieve visible results of improving the quality of garment generation by solving the problem of collapse.
2018-01-01T00:00:00ZConditional Adversarial Networks for Blind Image Deblurring
https://er.ucu.edu.ua:443/handle/1/1189
Conditional Adversarial Networks for Blind Image Deblurring
Kupyn, Orest
We present an end-to-end learning approach for motion deblurring,
which is based on conditional GAN and content loss – DeblurGAN.
DeblurGAN achieves state-of-the art in structural similarity measure
and by visual appearance. The quality of the deblurring model is also
evaluated in a novel way on a real-world problem – object detection
on (de-)blurred images. The method is 5 times faster than the closest
competitor.
Second, we present a novel method of generating synthetic motion
blurred images from the sharp ones, which allows realistic dataset
augmentation.
2018-01-01T00:00:00Z