Dissertationen zum Thema „Multi-view architectures“

Compte tenu de l'impact sévère et généralisé de la dépression, des initiatives de recherche significatives ont été entreprises pour définir des systèmes d'évaluation automatisée de la dépression. La recherche présentée dans cette thèse tourne autour des questions suivantes qui restent relativement inexplorées malgré leur pertinence dans le domaine de l'évaluation automatisée de la dépression : (1) le rôle de la structure du discours dans l'analyse de la santé mentale, (2) la pertinence de la représentation de l'entrée pour les capacités prédictives des modèles de réseaux neuronaux, et (3) l'importance de l'expertise du domaine dans la détection automatisée de la dépression.La nature dyadique des entretiens patient-thérapeute garantit la présence d'une structure sous-jacente complexe dans le discours. Dans cette thèse, nous établissons d'abord l'importance des questions du thérapeute dans l'entrée du modèle de réseau neuronal, avant de montrer qu'une combinaison séquentielle des entrées du patient et du thérapeute est une stratégie sous-optimale. Par conséquent, des architectures à vues multiples sont proposées comme moyen d'incorporer la structure du discours dans le processus d'apprentissage des réseaux neuronaux. Les résultats expérimentaux obtenus avec deux encodages de texte différents montrent les avantages des architectures multi-vues proposées, validant la pertinence de la conservation de la structure du discours dans le processus d'apprentissage du modèle.Ayant établi la nécessité de conserver la structure du discours dans le processus d'apprentissage, nous explorons plus avant les représentations textuelles basées sur les graphes. Les recherches menées dans ce contexte mettent en évidence l'impact des représentations d'entrée non seulement pour définir les capacités d'apprentissage du modèle, mais aussi pour comprendre leur processus prédictif. Les graphiques de similitude de phrases et les graphiques de corrélation de mots-clés sont utilisés pour illustrer la capacité des représentations graphiques à fournir des perspectives variées sur la même entrée, en mettant en évidence des informations qui peuvent non seulement améliorer les performances prédictives des modèles, mais aussi être pertinentes pour les professionnels de la santé. Le concept de vues multiples est également incorporé dans les deux structures graphiques afin de mettre en évidence les différences de perspectives entre le patient et le thérapeute au cours d'un même entretien. En outre, il est démontré que la visualisation des structures graphiques proposées peut fournir des informations précieuses indiquant des changements subtils dans le comportement du patient et du thérapeute, faisant allusion à l'état mental du patient.Enfin, nous soulignons le manque d'implication des professionnels de la santé dans le contexte de la détection automatique de la dépression basée sur des entretiens cliniques. Dans le cadre de cette thèse, des annotations cliniques de l'ensemble de données DAIC-WOZ ont été réalisées afin de fournir une ressource pour mener des recherches interdisciplinaires dans ce domaine. Des expériences sont définies pour étudier l'intégration des annotations cliniques dans les modèles de réseaux neuronaux appliqués à la tâche de prédiction au niveau des symptômes dans le domaine de la détection automatique de la dépression. En outre, les modèles proposés sont analysés dans le contexte des annotations cliniques afin d'établir une analogie entre leur processus prédictif et leurs tendances psychologiques et ceux des professionnels de la santé, ce qui constitue une étape vers l'établissement de ces modèles en tant qu'outils cliniques fiables
Given the severe and widespread impact of depression, significant research initiatives have been undertaken to define systems for automated depression assessment. The research presented in this dissertation revolves around the following questions that remain relatively unexplored despite their relevance within automated depression assessment domain; (1) the role of discourse structure in mental health analysis, (2) the relevance of input representation towards the predictive abilities of neural network models, and (3) the importance of domain expertise in automated depression detection.The dyadic nature of patient-therapist interviews ensures the presence of a complex underlying structure within the discourse. Within this thesis, we first establish the importance of therapist questions within the neural network model's input, before showing that a sequential combination of patient and therapist input is a sub-optimal strategy. Consequently, Multi-view architectures are proposed as a means of incorporating the discourse structure within the learning process of neural networks. Experimental results with two different text encodings show the advantages of the proposed multi-view architectures, validating the relevance of retaining discourse structure within the model's training process.Having established the need to retain the discourse structure within the learning process, we further explore graph based text representations. The research conducted in this context highlights the impact of input representations not only in defining the learning abilities of the model, but also in understanding their predictive process. Sentence Similarity Graphs and Keyword Correlation Graphs are used to exemplify the ability of graphical representations to provide varying perspectives of the same input, highlighting information that can not only improve the predictive performance of the models but can also be relevant for medical professionals. Multi-view concept is also incorporated within the two graph structures to further highlight the difference in the perspectives of the patient and the therapist within the same interview. Furthermore, it is shown that visualization of the proposed graph structures can provide valuable insights indicative of subtle changes in patient and therapist's behavior, hinting towards the mental state of the patient.Finally, we highlight the lack of involvement of medical professionals within the context of automated depression detection based on clinical interviews. As part of this thesis, clinical annotations of the DAIC-WOZ dataset were performed to provide a resource for conducting interdisciplinary research in this field. Experiments are defined to study the integration of the clinical annotations within the neural network models applied to symptom-level prediction task within the automated depression detection domain. Furthermore, the proposed models are analyzed in the context of the clinical annotations to analogize their predictive process and psychological tendencies with those of medical professionals, a step towards establishing them as reliable clinical tools

Thesis (Ph. D.)--Ohio State University, 2005.
Title from first page of PDF file. Document formatted into pages; contains xv, 128 p.; also includes graphics (some col.). Includes bibliographical references (p. 125-128). Available online via OhioLINK's ETD Center

Development of embedded systems poses an increasing challenge fordevelopers largely due to increasing complexity. Several factors contribute tothe complexity challenge: • the number of extra-functional properties applying to embedded systems,such as resource usage, timing effects, safety. • the functionality of embedded systems, to a larger extent than for othersoftware, involves engineers from multiple different disciplines, such asmechanical, control, software, safety, systems and electrical engineers.Themulti-disciplinarity causes the development environments to consistof separate data, models and tools. Several engineering paradigms to handle this complexity increase havebeen suggested, including methodologies focused on architecture, models andcomponents. In systems engineering, a long-standing approach has been todescribe the system in several views, each according to a certain viewpoint.By doing so, a divide-and-conquer strategy is applied to system concerns.Unfortunately, it is hard to always find completely independent concerns:there is always some semantic overlap between the different views. Modelbaseddesign (MBD) deals with building sound abstractions that can representa system under design and be used for analysis. Component-based design(CBD) focuses on how to build reusable component models with well-definedcomposition models. In this thesis, a concept of formalized multi-viewed component models (MVCM) is proposed, which integrates the three above mentioned paradigms.Principles and guidelines for MV CMs are developed. One of the main challengesfor the proposition is to provide MV CMs that produce composabilityboth along component boundaries and viewpoint boundaries. To accomplishthis, the relations between viewpoints need to be explicitly taken into account.Further, the semantic relations between these viewpoints need to be explicitlymodeled in order to efficiently ensure that the views are kept consistent. Asa main contribution, this thesis presents the formalization of the conceptsneeded to build such component models. A proper formalization of multiviewedconcerns provides several opportunities. Given suitable tool support, itwill be feasible to automate architecture analysis and architecture exploration. The thesis includes a number of case studies that provide insight andfeedback to the problem formulation and validating the results. The casestudies include a resource-aware reconfigurable middleware, a design of anarchitecture exploration methodology, and a windshield wiper system.

QC 20130527

Recent advancement in cameras and image processing technology has generated a paradigm shift from traditional 2D and 3D video to Multi-view Video (MVV) technology, while at the same time improving video quality and compression through standards such as High Efficiency video Coding (HEVC). In multi-view, cameras are placed in predetermined positions to capture the video from various views. Delivering such views with high quality over the Internet is a challenging prospect, as MVV traffic is several times larger than traditional video since it consists of multiple video sequences each captured from a different angle, requiring more bandwidth than single view video to transmit MVV. Also, the Internet is known to be prone to packet loss, delay, and bandwidth variation, which adversely affects MVV transmission. Another challenge is that end users’ devices have different capabilities in terms of computing power, display, and access link capacity, requiring MVV to be adapted to each user’s context. In this paper, we propose an HEVC Multi-View system using Dynamic Adaptive Streaming over HTTP (DASH) to overcome the above mentioned challenges. Our system uses an adaptive mechanism to adjust the video bitrate to the variations of bandwidth in best effort networks. We also propose a novel scalable way for the Multi-view video and Depth (MVD) content for 3D video in terms of the number of transmitted views. Our objective measurements show that our method of transmitting MVV content can maximize the perceptual quality of virtual views after the rendering and hence increase the user’s quality of experience.

This thesis is devoted to 2D and 3D modeling of urban environments using structured representations and grammars. Our approach introduces a semantic representation for buildings that encodes expected architectural constraints and is able to derive complex instances using fairly simple grammars. Furthermore, we propose two novel inference algorithms to parse images using such grammars. To this end, a steepest ascent hill climbing concept is considered to derive the grammar and the corresponding parameters from a single facade view. It combines the grammar constraints with the expected visual properties of the different architectural elements. Towards addressing more complex scenarios and incorporating 3D information, a second inference strategy based on evolutionary computational algorithms is adopted to optimize a two-component objective function introducing depth cues. The proposed framework was evaluated qualitatively and quantitatively on a benchmark of annotated facades, demonstrating robustness to challenging situations. Substantial improvement due to the strong grammatical context was shown in comparison to the performance of the same appearance models coupled with local priors. Therefore, our approach provides powerful techniques in response to increasing demand on large scale 3D modeling of real environments through compact, structured and semantic representations, while opening new perspectives for image understanding

The water's shore, used for defense, trade, transportation and industry, always plays an important role in a city's development. Buildings were built there to fulfill related practical functions, such as barns, factories or piers. However, at the present time, the functions mentioned above are not the only purpose of the shore, now more entertainment activies happen along the water's edge. Therefore, architecture's roles are changing. The site is in a small town called Riva San Vitale, which is located at the south end of Lake Lugano, Switzerland. For this small town, lake's shore is usually used by local people as a place for relaxation. Based on my four-month long living experiences there, I decided to design an art center on the lake to enrich residents' daily life.
Master of Architecture
A folding method was used in form processing to reflect the mountains' shapes surrounding the town. Through this whole process, my thesis explores how the architecture is used as an instrument to bring art into people's daily life and, at the same time, extend people's life into nature. As a result, the Art Center will become a new plaza of Riva San Vitale.

Thesis (S.M.)--Massachusetts Institute of Technology, School of Architecture and Planning, Program in Media Arts and Sciences, 2004.
Includes bibliographical references (leaves 71-75).
The number of three-dimensional displays available is escalating and yet the capturing devices for multiple view content are focused on either single camera precision rigs that are limited to stationary objects or the use of synthetically created animations. In this work we will use the existence of inexpensive digital CMOS cameras to explore a multi- image capture paradigm and the gathering of real world real-time data of active and static scenes. The capturing system can be developed and employed for a wide range of applications such as portrait-based images for multi-view facial recognition systems, hypostereo surgical training systems, and stereo surveillance by unmanned aerial vehicles. The system will be adaptable to capturing the correct stereo views based on the environmental scene and the desired three-dimensional display. Several issues explored by the system will include image calibration, geometric correction, the possibility of object tracking, and transfer of the array technology into other image capturing systems. These features provide the user more freedom to interact with their specific 3-D content while allowing the computer to take on the difficult role of stereoscopic cinematographer.
Samuel L. Hill.
S.M.

碩士
國立中正大學
資訊工程研究所
100
Nowadays, people require high quality video/image service in a more eager way. One movie or one ball game could not be filmed by only one camera. Multi cameras would be used simultaneously to get different view angles. As a result, there are much more video data needed to compress. If we adopt the past video coding standards, we need to process a laige amount of video data. The ITU-T formulates the Multi-view Video Coding (MVC) Standard which takes advantage of video data dependency from different views to reduce the quantity of data to be processed. The proposed thesis named “A View-Scalable Multi-view Decoder Architecture Design” is able to decode the video which contents different views of multi-view video and supports the view-scalable flexibility. With this feature, SoC designers could not only adjust the number of MVC decoder cores and set the firmware parameters according to system requirements. This feature also benefits to adopting 3D IC packaging or implementation by exploiting high bandwidth SDRAM access. In thsis proposed system, we are able to achieve two-view HD video decoding in real-time, so, we are able to use four decoder cores together for eight-view MVC video decoding. According to proposed design, we can achieve the demands of view-scalable multi-view decoder systems and make use of both the flexible stacking and high bandwidth access in 3D IC implementation. The proposed work is one feasible solution to approach flexible multi-view video decoder systems.

碩士
國立中正大學
資訊工程研究所
100
This thesis presents a low-bandwidth two-level inter-frame/inter-view cache architecture for a view scalable multi-view video decoder, which adopts two decoder cores to decode multi-view videos in parallel. The first level inter-frame cache is developed for the single video decoder core, which is able to reduce 40% bandwidth in P frame and 60% in B frame in doing inter-frame prediction in average. Moreover, we develop the second level inter-view cache architecture to reuse the same reference data for doing inter-view prediction among different decoder cores, which can further reduce 30% bandwidth. By adopting the proposed two-level cache architecture for doing inter-frame/inter-view prediction, we can reduce 80% bandwidth through a view scalable multi-view video decoder implementation, which achieves real-time HD1080 dual-view video decoding.

碩士
國立成功大學
電機工程學系
102
In order to offer immersive view experience for viewer, 3D stereoscopic technology applications are more popular. One of the most widely discussed topics is (depth image-based rendering) DIBR. The DIBR system renders virtual views at different viewpoints using 3D warping process. The DIBR calculates disparity value by depth value, and then moves the pixel of an object to generate new viewpoint image. During a 3D warping process, some parts of an occluded background region might be visible in a synthesized virtual image, i.e., a disoccluded region. It is referred to as a “hole”. The conventional way to solve this problem is using a smoothing filter to smooth depth map to reduce hole’s size before shifting, but it’s time-consuming. This thesis discards the traditional smoothing depth approach and proposes a novel depth map refinement scheme to remove blur pixels near the object boundary caused by restriction of camera and make the holes sit on the correct position. With the blur pixels removal, the hole-filling will become more reliable. The holes after warping process are filled by the proposed multi-direction hole-filling algorithms based on direction of edge. It can preserve the texture of background. The experimental results show that the visual quality of our proposal is competitive to others contemporary methods.

Thesis (M.Sc.)--York University, 2004. Graduate Programme in Computer Science.
Typescript. Includes bibliographical references (leaves 130-134). Also available on the Internet. MODE OF ACCESS via web browser by entering the following URL:http://gateway.proquest.com/openurl?url%5Fver=Z39.88-2004&res%5Fdat=xri:pqdiss&rft%5Fval%5Ffmt=info:ofi/fmt:kev:mtx:dissertation&rft%5Fdat=xri:pqdiss:MQ99311

碩士
國立臺灣大學
電子工程學研究所
103
With the rapid improvement of technology, light-field cameras have been growing more and more popular. Disparity map plays an important role for many light-field applications in light-field systems. Despite recent advances, state-of-the-art algorithms fail to generate a precise disparity map rapidly enough for VLSI real time processing. To generate real-time and accurate disparity map has become the bottleneck for current research. Disparity estimation can be formulated as an energy minimization problem on a 2D Markov Random Fields (MRFs). Among many MRF global optimization method, belief propagation (BP) is a popular global optimization algorithm which gives high quality and has several advantages for hardware implementation and highly potential to achieve real-time processing. However, it requires high bandwidth, memory and computational costs because of costly iterative operations, the original belief propagation is computationally expensive for real-time system implementation. Tile-based BP is an efficient improved BP for hardware implementation. Boundary messages loaded from different directions have been performed with different iterations to reduce high memory and bandwidth requirements. In this thesis, we first examine variant view configurations of light fields for multi-baseline disparity estimation for central view. We analyze the computation time of data cost generation against different number of views and the performance of each kind of variant light-field configurations for multi-baseline setting. After analysis on variant light-field configurations, we introduce current techniques of belief propagation and examine the problems of existing methods. Followed by introducing a novel hardware-efficient message-update method to remove redundant computational costs during message-update based on tile-based BP. We focus on the algorithm and hardware architecture design of multi-baseline disparity estimation from light fields. At first, we analyze the hardware cost in the belief propagation system, and indicate the challenge and bottleneck in the area and latency resource requirement to preserve high-quality in light fields. We analyze the computation time in the view of software implementation. After software analysis, we analyze the latency and area of the belief propagation from the perspective of hardware implementation. And then we exploit the unique characteristics of the generalized truncated linear model of the smoothness term in the Markov random field and propose an efficient hardware-oriented message-update algorithm from the aspect of hardware implementation with corresponding message-update process element (PE). This message-update PE indeed reduce the complexity of message construction for belief propagation, and hence greatly reduce the area and cycle counts.