- ItemOpen AccessGenerating 3D indoor environments with consistent styles(2022-11)In the current Master thesis, we address the scene generation task, with our main focus being on indoor scenes synthesis. Existing approaches pose the scene generation task as a layout creation problem. Namely the task is to populate a scene with a set of labelled bounding boxes that correspond to a set of furniture pieces. In particular, these methods typically seek to learn a probability distribution over a set of attributes that define each object such as their shape, category, orientation and position in the scene. During generation, the generated bounding boxes are replaced with 3D objects by retrieving them from a library of assets based on various criteria such as size, object category etc. Naturally, since the object retrieval process is independent from the layout generation there are no guarantees that the generated objects will be coherent in terms of style and appearance. To this end, in this work, we propose a novel generative model for indoor scenes that takes into consideration the per-object style during the generation process. We believe that this is a crucial step towards generating realistic environments. In particular, we build on top of ATISS , which is the state-of-the-art indoor scene generation pipeline. Specifically, we extend its capabilities by also incorporating a style prediction module. Further- more, we also propose a novel retrieval procedure that instead of simply relying on the size to replace bounding boxes with 3D models, takes into account the per-object style. Our experimental evaluation showcases that our model consistently generates stylistically meaningful scenes, (i.e. the nightstands next two a bed or the chairs around the table should have similar appearance), while performing on par with ATISS wrt. the scene generation quality. Finally, we also introduce various metrics that can be used for evaluating the generated scenes in terms of the style coherence.
- ItemOpen AccessReconstructing surfaces with appearance using neural implicit 3D representations(2022-11-15)Parsing complex 3D scenes into compact low-dimensional representations has been a long-standing goal in Computer Vision that could tremendously benefit various downstream applications such as scene understanding and reconstruction. Based on the output of the final representation, existing methods can be categorized to explicit and implicit methods. In particular, implicit-based approaches have recently gained popularity due to their simple yet efficient parameterization. The primary goal of these works (such as OccNet, SRN, Neural Volumes etc,) is to create implicit representations by mapping 3D points with the pertinent scenes. Although these techniques render promising results they struggle to learn to represent complex scenes. Neural Radiance Fields was the breakthrough in this direction. Mapping scene geometry and appearance with a spatial 3D location turned out to be the simplest and most effective method since then. Not only did it overpass the previous methods in terms of fidelity and accuracy, but was able to encode even complex scenes. Neural Radiance Fields as originally proposed by Mildenhall et al. are limited to only single scene overfitting and time efficiency. Since mapping is performed by training an MLP for a scene specifically, many methods proposed afterwards came to address this problem. GRF and PixelNeRF for example exploit image features in order to learn scene priors that allow multi scene training. Despite the fact that NeRF produced state of the art results in novel view synthesis task, it lacked accuracy in terms of 3D reconstruction. Recent works such as UniSURF show that we can produce accurate surface reconstructions by combining surface and volumetric rendering. In our work, we combine PixelNeRF with UniSURF, by applying accurate surface extraction methods to multi scene 3D implicit representations.
- ItemOpen AccessA nonnegative least squares solver for mjultiple right-hand sides for approximating the nonnegative matrix factorization(2022-11-04)Nonnegative Least Squares (NNLS) problems, where the variables are restricted to take only nonnegative values, often arise in many applications and are also at the core of most approaches to solve the nonnegative matrix factorization (NMF), a low-rank matrix approximation problem with nonnegativity constraints. NMF is a data analysis technique used in a great variety of applications such as text mining, image processing, hyperspectral data analysis, computational biology, and clustering. In more detail, the nonnegative factors can be interpreted as data e.g., as images described by pixel intensities or texts represented by vectors of word counts. The mathematical formulation for NMF appears as a non-convex optimization problem, and various types of algorithms have been devised to solve the problem. The first goal of this thesis is to propose a new eﬀicient, yet simple to implement, approach to solve nonnegative linear least squares problems for multiple right-hand sides. More precisely, we study and use properties of global algorithms for least squares problems which are then combined with rules that enforce nonnegativity and lead to novel techniques for solving the aforementioned problem by a flexible Krylov subspace method. Comparisons of the state of the art algorithms using datasets and examples that come from real life applications as well as those artificially generated show that the proposed new algorithm presents a satisfactory behaviour and in some cases outperforms existing ones in computational speed and accuracy. Our second goal is to study extensively the NMF, its properties and applications and dive into the existing algorithms and methodologies used in order to approximate a solution for it. Moreover, using our new approach, tuned to solve large scale nonnegative least squares problems for multiple right-hand sides we present a novel algorithm for NMF based on the alternating nonnegative least squares (ANLS) framework. Extensive experiments on document clustering, images and synthetic datasets indicate the effectiveness of our approach.
- ItemEmbargoStudy and optimization of digital filters transition(2022)The center of this thesis is digital filter transition; a problem encountered in audio applications, such as speech coding, equalization of audio signal and music synthesis. Those are some cases, where filter switching can cause undesirable sound effects (clicks, plops, gaps etc). In the first part of the present thesis, an extensive survey of existing transition techniques is presented, covering a large part of current literature. After a selection of widely-known methods from the aforementioned review, we implemented and tested the respective algorithms in MATLAB environment. To assess the implementation results, objective evaluation is performed through certain commonly accepted models and metrics. Based on this evaluation, the optimal transition algorithm is identified and implemented in real time, on a digital signal processor (DSP).
- ItemEmbargoReconfigurable intelligent surfaces for applications in 6G wireless systems(2022)Moving from fifth-generation (5G) into Beyond 5G (B5G) and sixth-generation (6G) wireless networks, both academia and industry around the world have already started to investigate advanced technologies. More stringent requirements such as ultra high data rate, high energy-efficiency, extremely high reliability, ultra low latency, global coverage and connectivity are considered in future networks. Reassessment of key performance indicators (KPIs) and definition of new KPIs in future use cases are required, since not only novel 6G technologies are introduced but also changes in the architecture of the conventional 5G cellular networks are expected. Massive multiple-input-multiple-output (MIMO), the main technology of 5G, is a reality and research community investigates new promising physical layer technologies. Although massive MIMO resolves many basic problems of wireless communications, several open problems remain. The usage of multiple antennas enables beamforming, thus increasing the received signal power and liming interference issue at the end user. However, this does not apply for each served user. Large signal variations are observed in a real system especially for cell-edge users, due to possible blockages that are placed between the communication link. Furthermore, interference problem arises from other neighbor cells. Apart from performance metrics related to the user quality of experience, a base station (BS) equipped with a massive number of antennas features with high power consumption, that comes into conflict with the energy-efficient future networks. Several technologies such as Reconfigurable Intelligent Surfaces (RISs), Cell-Free MIMO and Orthogonal Time Frequency Spatial (OTFS) modulation are emerging in 6G aiming to solve these open problems. RISs are envisioned as a new physical layer technology in 6G wireless communication systems. A RIS is a two-dimensional, low-cost surface that is strategically placed in the infrastructure and introduces an optimal phase shift to the incident signal, thus reflecting the electromagnetic wave in the desired direction of the user. Therefore, RIS installation can be used to achieve smart propagation environments by turning the wireless channel from a passive actor into a service. Properties of the meta-surface can be altered and controlled in real-time, while low power consumption is the main benefit of RIS node compared to other physical layer technologies. These features explain the great interest of the research community on this novel technology. Various possible use cases are presented in this work, proving the potential of RIS technology in future wireless communication networks. In this thesis, we study the efficiency of power iteration-based methods in the received signal power maximization problem, assuming a RIS-aided Orthogonal Frequency Division Multiplexing (OFDM) system. Apart from state-of-the-art passive beamforming methods, general techniques that solve the uni-modular quadratic program (UQP) problem, are also examined. Binary phase shifts with unbalanced amplitudes and mutual coupling are considered, while a method that is easily implemented on hardware is proposed. Furthermore, proposed architectural optimizations, related to memory and loop dependencies management, lead to a low latency-oriented solution. Thus, the implementation of the proposed method, on a Zynq UltraScale+ multiprocessor system on a chip (MPSoC) device, results in an extremely low execution time.