| Topic title |
Possible scientific supervisors |
Source of funding |
|
Generative adversarial network usage in motion matching
|
prof. dr. Rytis MASKELIŪNAS |
state-funded |
Research Topic Summary.
In current digital era, animations and their usage are an integral part of digital entertainment. Also, motion blending is one of the areas little touched by animation. Research on how deep neural networks could contribute to motion-tuning of animations remains relevant. The aim of the research is to create an algorithm that would be able to dynamically adapt and adjust motion animations to a specific situation, by analyzing various features on the fly, like an object’s reaction to force, motion or other vectors in a certain matrix, thus distancing itself from the limitations found in a classic fixed animations. Prerequisites: Artificial Intelligence, Object Oriented Programming, animation methods, game technologies.
|
|
Deep learning based Dynamic motion synthesis for on-the-fly skeleton animation
|
prof. dr. Rytis MASKELIŪNAS |
state-funded |
Research Topic Summary.
The main goal of this topic is a deep learning based dynamic motion generation system based on motion fields, for responsive and accurate real-time skeleton animation. This topic’s novelty is a focus on a method to generate enhanced animated movements more naturally and interactively by utilizing "on-the-fly" programming animation, combining artificial intelligence, biomechanics, physics and numerous other parameters available during run-time. Prerequisites: Artificial Intelligence, Object Oriented Programming, animation programming.
|
|
Deep neural network-based method for detecting mechanical failures in noisy acoustic environments
|
prof. dr. Rytis MASKELIŪNAS |
state-funded |
Research Topic Summary.
Factory machinery is prone to failure or breakdown and often this leads to mechanical failures not detectable by a set of common sensors. This thesis hypothesizes on an idea of distinguishing mechanical failures from acoustic features. The novelty aspect comes from understanding that the environment is often very noisy, which requires development of specific intelligent methods for sound data-based anomaly detection of industrial machinery. Prerequisites: Artificial Intelligence, Object Oriented Programming, signal processing.
|
|
Posture detection method from a limited video material
|
prof. dr. Rytis MASKELIŪNAS |
state-funded |
Research Topic Summary.
The idea of this thesis is to investigate a possibility to investigate a method capable of reducing or eliminating posture detection problems related to a limited visibility human torso in the frame, i.e., the occlusion problem, as well as dependencies on camera angles, signal clarity and environmental issues such as lighting. Novelty comes from current technological limitations, where current skeleton-based posture prediction and detection methods fail due to the reason of not enough data received by the sensors to infer the full human skeleton or distortions limiting the recreation of that model. Prerequisites: Artificial Intelligence, Object Oriented Programming, signal processing technologies.
|
| Framework for Specifying and Evaluating Usability during Gamified System Development |
prof. dr. Rimantas BUTLERIS |
state-funded |
|
Microbial process modeling and optimal control
|
prof. dr. Renaldas URNIEŽIUS |
state-funded |
Research Topic Summary.
Biotechnological processes are among the most complicated control objects due to nonlinear relationships between the process variables, changing of process dynamic parameters in wide ranges during fed-batch cultivation processes, lack in reliable sensors for direct measurement of important state variables. Therefore, development of effective soft sensor systems is a relevant bio-engineering task. Demand for such systems is high, they are relevant for development of new and improvement of existing bio-processes both in research laboratories and industrial companies. The developed biotechnological process monitoring and optimal control algorithms and systems would represent new results in bio-process state estimation and control practice. New knowledge in the soft sensors development and applications will be obtained and the software technologies for realization of the soft sensor systems will be developed and tested. KTU Dept. of Automation is currently performing a scientific research grant “Development and application of soft sensor system for state estimation of biotechnological processes” (No. 01.2.2- LMT-K-718-03-0039), under which there are opportunities for new researchers to simultaneously perform their research and participate in the compensated application development for biopharmaceutical industry.
|
|
A study of Virtual Reality application for learning systems
|
prof. dr. Tomas BLAŽAUSKAS |
state-funded |
Research Topic Summary.
A large proportion of people have a dominant kinesthetic sensory preference. Besides, human learning depends on the environment around them. The emergence of high-quality and relatively inexpensive virtual reality devices, such as HTC Vive, Oculus Rift, Oculus Go, etc., has made it possible to realistically simulate the surrounding environment and simulate real-world activities, thus opening up opportunities for kinesthetic distance learning. The objectives of the research are: to analyze the methods of using virtual reality for learning; to develop the concept of a learning system that incorporates virtual reality; to propose a model of a multimodal system; to offer ways of providing lectures in virtual reality and ways to learn by using it; evaluate the effectiveness of using virtual reality for learning. The expected results are the proposed concepts, models, and practical implementations used for learning using virtual reality and the experimental evaluation of the proposed solutions.
|
|
Cryptographic Algorithms Specification Methods for Power Consumption Forecasting and Management in Mobile Devices
|
prof. dr. Jevgenijus TOLDINAS |
state-funded |
Research Topic Summary.
The number of mobile phone users in the world is expected to pass the five billion mark by 2019. Today mobile phone in our pocket is a high performance computer with gigabytes of RAM, processor that has two, four or more cores, Wi-Fi, Bluetooth, display, etc. All of mentioned devices consume battery energy. Unfortunately, while technical parameters of mobile phones growths quickly, battery lifetime remains very short and users must charge their phones more and more frequently. BYOD (bring your own device) is the increasing trend when employee brings their owned smartphones, tablets or laptops to their workplace. That is why prediction of energy consumption for devices with battery is an important issue. Main purpose of the dissertation: 1) To propose novel models, methods and software tools for energy consumption forecasting at application level; 2) To specify how much cryptographic algorithms consume the power of a mobile device. New results are expected: 1) Common models, specialized methods that assess the variability of the criteria and enable prediction, management of energy consumption at the application level; 2) Realization of specialized methods - software components that enable users to evaluate the energy consumption of cryptographic algorithms in relation to other criteria; (3) Adaptation of specialized methods for the flexible use of mobile devices for energy forecasting.
|
| Preventive system maintenance using process mining and anomaly detection |
prof. dr. Evaldas VAIČIUKYNAS |
state-funded |
|
Machine learning models for solving object shape recognition problems
|
doc. dr. Armantas OSTREIKA |
state-funded |
Research Topic Summary.
When applying artificial intelligence algorithms, it is important to properly select criteria that combine the essential properties of the object or phenomenon under consideration, and thus to create an appropriate criteria/parameter model that allows unambiguous identification and classification of object state changes. In computer vision tasks, the creation of such models is not a trivial task, as visual information is enormously diverse and, moreover, changes dynamically. Properly designed and implemented, such models serve for high-level decision-making using a digital image or video data. These solutions are designed to automate tasks that can only be performed by a human. Computer vision systems are used in many fields, including manufacturing, medical, traffic monitoring, security systems, etc. Computer vision methods allow automatic analysis of important information in the processing of large data flows. Artificial intelligence algorithms in this case are an integral part of solving computer vision problems. The proposed topic will aim to improve existing common methodologies by focusing on specific products and real-world processes for greater efficiency and reliability, adapting and improving machine learning methods in Decision trees, K nearest neighbors, Naïve Bayes, Support vector classifiers, Direct propagation neural networks, and Convolutional neural networks. For more details please contact the supervisor of the topic.
|
