FITAbbreviation: DVI4Acad. year: 2020/2021
Programme: Computer Science and Engineering
Length of Study: 4 years
Tuition Fees: 4000 EUR/academic year for EU students, 4000 EUR/academic year for non-EU students
Accredited from: 1.1.2007Accredited until: 31.12.2024
The goal of the doctoral study programme is to provide outstanding graduates from the MSc study programme with a specialised university education of the highest level in certain fields of information technology, including especially the areas of information systems, computer-based systems and computer networks, computer graphics and multimedia, and intelligent systems. The education obtained within this study programme also comprises a training and attestation for scientific work.
prof. RNDr. Milan Češka, CSc.
Issued topics of Doctoral Study Program
The topic concerns algorithms of computer graphics and image synthesis. Its main goal is to research new algorithms so that their features and application possibilities are better understood so that they are improved or newly created. The programming work is expected in C, C++, C#, assembly language, or other languages. If suitable, they can be efficiently implemented e.g. in CPU, such as x86/64, ARM, Xeon PHI, GPU, etc. or other cores in CUDA, OpenCl, VHDL, etc. Algorithms of interest include:
After mutual agreement, individually selected algorithms can be considered as well as soon as they do belong to the general topic.
Collaboration on grant projects, such as TACR, H2020, ECSEL possible (employment or scholarship).
Tutor: Zemčík Pavel, prof. Dr. Ing., dr. h. c.
The project is concerned with advanced methods of computational photography. The aim is to research new computational photography methods, which comprises software solutions potentially supported by new optics and/or hardware. Our interest is on HDR image and video processing, color-to-grayscale conversions, spectral imaging, and others.
Tutor: Čadík Martin, doc. Ing., Ph.D.
The project is concerned with advanced rendering and global illumination methods. The aim is to research new photorealistic (physically accurate) as well as non-photorealistic (NPR) simulations of interaction of light with the 3D scene. Cooperation and research visits with leading research labs are possible (Adobe, USA, MPII Saarbrücken, Německo, Disney Curych, Švýcarsko, INRIA Bordeaux, Francie).
Trusted computing is a promising concept that enable remote trusted execution of arbitrary programming logic. The most prominent examples of trusted execution environments (TEE) are Intel SGX, IBM Sanctum, Keystone-enclave. The goal of this work is to analyze all existing TEEs, they properties and potential applications. Then, the work should propose novel applications of TEEs for the problems that were not resolved before while also implement the most interesting application. There is a high potential to combine trusted computing with public blockchains, which should be also addresses in the work.
Tutor: Hanáček Petr, doc. Dr. Ing.
The purpose of this thesis is to analyze potential application of cryptographic accumulators in the context of the public blockchains. Identify the most suitable types and settings for particular application and also propose improvements that focus on the performance, security, or storage overhead.
Téma práce se zabývá využitím technik dolování informací z řeči jako je rozpoznávání mluvčího, rozpoznávání jazyk, přepis, a detekce klíčových slov pro vyhledávání struktur zločineckých sítí. Téma souvisí s evropským projektem Horizon 2020 Roxanne, kde je VUT hlavním výzkumným partnerem.
Zadání vyžaduje zájem o matematiku, statistiku, strojové učení a zpracování řeči, výhodou je zběhlost v jazyce Python a jeho knihovnách pro strojové učení.
BUT Speech@FIT je špičkovou mezinárodní výzkumnou skupinou zabývající se dolováním informací z řeči. Nabízí rovné příležitosti pro všechny, v současné době má členy 10ti národností a (na IT) významný podíl žen.
Více informací viz https://www.roxanne-euproject.org/ a https://speech.fit.vutbr.cz/
Tutor: Černocký Jan, prof. Dr. Ing.
Photoacoustic Tomography (PAT) is an emerging 3D biomedical imaging modality for both clinical and pre-clinical imaging that has gained increasing attention in the last decade. In principle, PAT can achieve as high spatial and temporal resolution as ultrasound images, but, because it depends on optical absorption for contrast, it also has the potential to provide functional, molecular and genetic imaging capabilities. However, realising these advantages requires high quality image reconstruction algorithms.
Several approaches to PAT image reconstruction are currently used, including time reversal, series constructions, and filtered backprojection algorithms. However, there are several aspects that cannot be taken into account in these classical approaches and therefore result in image artefacts, including (1) patient movement, (2) incomplete coverage of the object with ultrasound sensors, (3) angle and frequency variability of the sensor's sensitivity, (4) unstable environment, e.g., variations in temperature, (5) sound speed variations with the object eg. breast, and (6) measurement noise. In addition, large-scale 3D images take time to reconstruct, limiting the imaging frame rate; increasing the frame rate would open up many more potential applications.
Currently, the reconstruction of photoacoustic images takes several hours on high-end servers and subsequently artefacts are detected and removed. The goal of this thesis is to overcome some of these limitations, by improving reconstruction speed and quality of photoacoustic images using artificial intelligence.
We can identify three essential approaches for this task. First, use AI to correct or complement the measured data to form a complete and clean data set then use a classical reconstruction algorithm. Second, use AI to complement the image reconstruction task, by intertwining learnable components with classical reconstruction approaches. Third, use a classical reconstruction algorithm to form an image and then use AI to remove the artefacts.
These topic cover several areas, from the experimental acquisition of photoacoustic data, techniques for image reconstruction and high performance computing, to deep learning and artificial intelligence.
Tutor: Jaroš Jiří, doc. Ing., Ph.D.
The goal of the work is to research and create algorithms that will allow for running augmented reality on mobile (ultramobile) devices. It mainly concerns algorithms of pose estimation in the space by the means of computer vision and by using sensors embedded in the device. Furthermore, the work will elaborate on algorithms of rendering of virtual elements into the real-world scene and on applications of augmented reality on mobile devices.
Tutor: Herout Adam, prof. Ing., Ph.D.
The aim of the thesis is to create 3D face model from 2D photos of diverse origins, namely:
Tutor: Drahanský Martin, prof. Ing., Ph.D.
Deep convolution networks have been a clear trend of machine learning for image analysis in recent years. Neural networks can also be used for 3D image analysis, where the network works with 3D convolutions. However, this approach is problematic because of its huge memory and computational requirements.The aim of the dissertation thesis is to explore, analyze and design new architectures of deep neural networks and approaches to their learning for 3D object shape recognition tasks, where the data set consists of various 3D data representations - e.g. 3D meshes, voxel representation, etc.Proposed methods will be applied in the projects on which the supervisor participates.
Tutor: Španěl Michal, Ing., Ph.D.
Deep convolution networks have been a clear trend of machine learning for image analysis in recent years. However, in tasks with a very small and specific data set, where it is not enough to use data augmentation or GAN concepts, their usage is still problematic.The goal of the dissertation thesis is to explore, analyze and design new architectures of deep convolutional networks and approaches to their learning for image analysis tasks in which the size of the annotated data set is extremely small or is gradually growing. For learning neural networks it is possible to use unannotated data or partially annotated data in the form of a limited user input.Proposed methods will be applied in the projects on which the supervisor participates.
The topic focuses embedded image, video and/or signal processing. Its main goal is to research capabilities of "smart" and "small" units that have such features that allow for their applications requiring smyll, hidden, distributed, low power, mechanically or climatically stressed systems suitable of processing of some signal input. Exploitation of such systems is perspective and wide and also client/server and/or cloud systems. The units themselves can be based on CPU/DSP/GPU, programmable hardware, or their combination. Smart cameras can be considered as well. Applications of interest include:
A possibility exists in collaboration on grant projects, especially the newly submitted TAČR, H2020, ECSEL ones (potentially employment or scholarship possible).
The aim of this work is to generate various damages into synthetic fingerprints and analysis of their quality. The work will consist of:
The goal of this thesis is to analyze side-channels and software vulnerabilities in hardware wallets, which are currently considered as a most secure way of storing private keys of users. These hardware wallets are connected to the client machine by a USB, Bluetooth, or other connection. Therefore, we assume two attacker models, one has the physical access to the wallet and another one tampers with the client interface and thus can influence the execution of the client protocol.Examples of wallets that we want to analyze are Trezor One/T, CoolBitX, Ellipal, Ledger Nano S, etc. Hardware skills (oscilloscope) are advantage for the PhD student.
The project deals with image and video quality assessment metrics (IQM). The aim is to explore new ways how to incorporate human visual system properties into IQM. In particular, we will consider perception of HDR images, and utilization of additional knowledge (in form of metadata, 3D information, etc.) about the tested scenes using machine learning (e.g. neural networks).
The project is concerned with advanced methods of image processing. The aim is to research new methods using machine learning, in particular deep convolutional neural networks.
The topic of identifying and extracting specific information from documents on the Web has been the subject of intensive research for quite a long time. The basic obstacles that make this problem difficult are the loose structure of HTML documents and absence of meta-information (annotations) useful for recognizing the content semantics. This missing information is therefore compensated by the analysis of various aspects of web documents that include especially the following:
A background knowledge about the target domain and the commonly used presentation patterns is also necessary for successful information extraction. This knowledge allows a more precise recognition of the individual information fields in the document body.
Current approaches to information extraction from web documents focus mainly modeling and analyzing the documents themselves; modeling the target information for more precise recognition has not yet been examined in detail in this context. The assumed goals of the dissertation are therefore the following:
Experimental implementation of the proposed methods using existing tools and experimental evaluation on real-world documents available on the WWW is also an integral part of the solution.
Tutor: Burget Radek, doc. Ing., Ph.D.
The aim of this dissertation is to design and implement software for evaluation of measured data of cylindrical cavities. Based on the measured data, the application creates a computer model of the cavity in which it will look for damaged or otherwise problematic sites. The system will be able to predict the development of this site in the future under the same conditions of use of a device containing a measured cylindrical cavity (e.g. a barrel of military weapons). The device is available + can be fitted with other sensors as required. Research can be divided into the following stages:
Current deep machine learning methods are based on continuous vector representations that are created by the neural networks (NN) themselves during the training. Although empirically, the results of NNs are often excellent, ourknowledge and understanding of such representations is insufficient. The task for this dissertation is to study neural representations for speech and text units of different scopes (from phonemes and letters to whole spoken and written documents) and representations acquired both for isolated tasks and multi-task setups:
Tutor: Burget Lukáš, doc. Ing., Ph.D.
The topic concerns algorithms of image, video, and/or signal processing. Its main goal is to research and in-depth analyze existing algorithms and discover new ones so that they have desirable features and so that they are possible to efficiently implement. Such efficient implementation can be but does not necessarily have to be part of the work but it is important to prepare the algorithms so that they can be efficiently implemented e.g. in CPU, in CPU with acceleration through SSE instructions, in embeded systems, in embedded systems with FPGA, in Intel Xeon PHI, in extremely low power systems, or in other environments. The programming work is expected in C, C++, C#, assembly language, CUDA, OpenCl, VHDL, or other languages. The application possibilities of the algorithms are also important and the application can be but does not have to be part of the work. The algorithms/applications of interest include:
Safety, security, privacy and environmental protection are of paramount importance to respective unmanned aircraft systems (UAS) regulatory agencies. The widespread acceptance of UAS operations depends on their ability to demonstrate compliance with domestic, regional and international regulations and policies.The present research/thesis aims to identify the technical, management and regulatory challenges facing by the UAS operation. Based on the gap analysis, the research/thesis wil try to give one big-data driven UAS Traffic Management (UTM) solution. The design of UTM will also consider the interaction and integration with the current and future Air Traffic Management(ATM) system.To make this research/thesis possible, we may introduce the resources from world-leading organizations, for example, ICAO and JARUS, and also their leaders and experts.
Tutor: Chudý Peter, doc. Ing., Ph.D., MBA
The aim of the thesis is reconstruction of damaged CD/DVD/BR/HDD surfaces, consisting of:
The aim of this work is to create a reliable matching of 2D face with 3D face projection, namely:
Semantic web technology allows the representation of information and knowledge for the purpose of its further sharing, for example, in computer applications. Available knowledge databases, such as DBPedia, contain a great amount of useful information and facts. On the current web, however, most of the new information is published in the form of documents most often in HTML, whose further processing is problematic mainly due to their free structure and the absence of explicit information about the meaning of individual parts of the content. There exist two ways to overcome this gap between the classical and the semantic web:
To achieve both these goals, it is necessary to analyze the capabilities of existing ontological models regarding the modelling of the target domains and mapping these descriptions on the content of real-world web pages and documents. Possible applications include, but are not limited to:
Experimental implementation of the proposed methods using the existing tools and experimental evaluation on real-world data and documents is also an integral part of the solution.
The advent of autonomous mobility leverages the importance of vehicle's safe collision avoidance and an elimination of the need for human monitoring. The aim of the thesis is to research an integrated opto-radio-electronic system capable of issuing alert due to nearby traffic and a real-time estimation of the collision objects' trajectories. With the increase of the Unmanned Aircraft Systems' (UAS) popularity, a reliable detection and collision avoidance becomes a necessity for autonomous UAS operation beyond visual line of sight conditions.
The project deals with geo-localization of mobile devices in unknown environments using computer vision and computer graphics methods. The aim is to investigate and develop new image registration techniques (with geo-localized image database or 3D terrain model). The goal is an efficient implementation of proposed methods on mobile devices as well as search for additional applications in the area of image processing, computational photography, and augmented reality.