Tesis sobre el tema "Video compression. Real-time data processing"

Thesis (M. Phil.)--Hong Kong University of Science and Technology, 2004.
Includes bibliographical references (leaves 205-212). Also available in electronic version. Access restricted to campus users.

Après l'entrée dans l'ère du numérique, la consommation vidéo a évolué définissant de nouvelles tendances. Les contenus vidéo sont désormais accessibles sur de nombreuses plateformes (télévision, ordinateur, tablette, smartphone ... ) et par de nombreux moyens, comme les réseaux mobiles, les réseaux satellites, les réseaux terrestres, Internet ou le stockage Blu-ray par exemple. Parallèlement, l'expérience utilisateur s'améliore grâce à la définition de nouveaux formats comme l'Ultra Haute Définition (UHD), le « High Dynamic Range » (HDR) ou le « High Frame Rate » (HFR). Ces formats considèrent une augmentation respectivement de la résolution, de la dynamique des couleurs et de la fréquence d'image. Les nouvelles tendances de consommation et les améliorations des formats imposent de nouvelles contraintes auxquelles doivent répondre les codeurs vidéo actuels et futurs. Dans ce contexte, nous proposons une solution de codage vidéo permettant de répondre à des contraintes de codage multi-formats, multi-destinations, rapide et efficace en termes de compression. Cette solution s'appuie sur l'extension Scalable du standard de compression vidéo « High Efficiency Video Coding » (HEVC) définie en fin d'année 2014, aussi appelée SHVC. Elle permet de réaliser des codages scalables en produisant un unique bitstream à partir d'un codage sur plusieurs couches construites à partir d'une même vidéo à différentes échelles de résolutions, fréquences, niveaux de qualité, profondeurs des pixels ou espaces de couleur. Le codage SHVC améliore l'efficacité du codage HEVC grâce à une prédiction inter-couches qui consistent à employer les informations de codage issues des couches les plus basses. La solution proposée dans cette thèse s'appuie sur un codeur HEVC professionnel développé par la société Ateme qui intègre plusieurs niveaux de parallélisme (inter-images, intra-images, inter-blocs et inter-opérations) grâce à une architecture en pipeline. Deux instances parallèles de ce codeur sont synchronisées via un décalage inter-pipelines afin de réaliser une prédiction inter-couches. Des compromis entre complexité et efficacité de codage sont effectués au sein de cette prédiction au niveau des types d'image et des outils de prédiction. Dans un cadre de diffusion, par exemple, la prédiction inter-couches est effectuée sur les textures pour une image sur deux. A qualité constante, ceci permet d'économiser 18.5% du débit pour une perte de seulement 2% de la vitesse par rapport à un codage HEVC. L'architecture employée permet alors de réaliser tous les types de scalabilité supportés par l'extension SHVC. De plus, pour une scalabilité en résolution, nous proposons un filtre de sous-échantillonnage, effectué sur la couche de base, qui optimise le coût en débit global. Nous proposons des modes de qualité intégrant plusieurs niveaux de parallélisme et optimisations à bas niveau qui permettent de réaliser des codages en temps-réel sur des formats UHD. La solution proposée a été intégrée dans une chaîne de diffusion vidéo temps-réel et montrée lors de plusieurs salons, conférences et meetinqs ATSC 3.0
After entering the digital era, video consumption evolved and defined new trends. Video contents can now be accessed with many platforms (television, computer, tablet, smartphones ... ) and from many medias such as mobile network or satellite network or terrestrial network or Internet or local storage on Blu-ray disc for instance. In the meantime, users experience improves thanks to new video format such as Ultra High Definition (UHD) or High Dynamic Range (HOR) or High Frame Rate (HFR). These formats respectively enhance quality through resolution, dynamic range and frequency. New consumption trends and new video formats define new constraints that have to be resolved by currents and futures video encoders. In this context, we propose a video coding solution able to answer constraints such as multi-formats coding, multi­destinations coding, coding speed and coding efficiency in terms of video compression. This solution relies on the scalable extension of the standard « High Efficiency Video Coding » (HEVC) defined in 2014 also called SHVC. This extension enables scalable video coding by producing a single bitstream on several layers built from a common video at different scales of resolution, frequency, quality, bit depth per pixel or even colour gamut. SHVC coding enhance HEVC coding thanks to an inter-layer prediction that use coding information from lower layers. In this PhD thesis, the proposed solution is based on a professional video encoder, developed by Ateme company, able to perform parallelism on several levels (inter-frames, intra-frames, inter-blocks, inter-operations) thanks to a pipelined architecture. Two instances of this encoder run in parallel and are synchronised at pipeline level to enable inter-layer predictions. Some trade-off between complexity and coding efficiency are proposed on inter-layer prediction at slice and prediction tools levels. For instance, in a broadcast configuration, inter-layer prediction is processed on reconstructed pictures only for half the frames of the bitstream. In a constant quality configuration, it enables to save 18.5% of the coding bitrate for only 2% loss in terms of coding speed compared to equivalent HEVC coding. The proposed architecture is also able to perform all kinds of scalability supported in the SHVC extension. Moreover, in spatial scalability, we propose a down-sampling filter processed on the base layer that optimized global coding bitrate. We propose several quality modes with parallelism on several levels and low-level optimization that enable real-time video coding on UHD sequences. The proposed solution was integrated in a video broadcast chain and showed in several professional shows, conferences and at ATSC 3.0 meetings

With heightened security concerns across the globe and the increasing need to monitor, preserve and protect infrastructure and public spaces to ensure proper operation, quality assurance and safety, numerous video cameras have been deployed. Accordingly, they also need to be monitored effectively and efficiently. However, relying on human operators to constantly monitor all the video streams is not scalable or cost effective. Humans can become subjective, fatigued, even exhibit bias and it is difficult to maintain high levels of vigilance when capturing, searching and recognizing events that occur infrequently or in isolation. These limitations are addressed in the Live Video Database Management System (LVDBMS), a framework for managing and processing live motion imagery data. It enables rapid development of video surveillance software much like traditional database applications are developed today. Such developed video stream processing applications and ad hoc queries are able to "reuse" advanced image processing techniques that have been developed. This results in lower software development and maintenance costs. Furthermore, the LVDBMS can be intensively tested to ensure consistent quality across all associated video database applications. Its intrinsic privacy framework facilitates a formalized approach to the specification and enforcement of verifiable privacy policies. This is an important step towards enabling a general privacy certification for video surveillance systems by leveraging a standardized privacy specification language. With the potential to impact many important fields ranging from security and assembly line monitoring to wildlife studies and the environment, the broader impact of this work is clear. The privacy framework protects the general public from abusive use of surveillance technology; success in addressing the “trust” issue will enable many new surveillance-related applications. Although this research focuses on video surveillance, the proposed framework has the potential to support many video-based analytical applications.
Ph.D.
Doctorate
Computer Science
Engineering and Computer Science
Computer Science

Runway detection in long wavelength infrared (LWIR) video could potentially increase the number of successful landings by increasing the situational awareness of pilots and verifying a correct approach. A method for detecting runways in LWIR video was therefore proposed and evaluated for robustness, speed and FPGA acceleration. The proposed algorithm improves the detection probability by making assumptions of the runway appearance during approach, as well as by using a modified Hough line transform and a symmetric search of peaks in the accumulator that is returned by the Hough line transform. A video chain was implemented on a Xilinx ZC702 Development card with input and output via HDMI through an expansion card. The video frames were buffered to RAM, and the detection algorithm ran on the CPU, which however did not meet the real-time requirement. Strategies were proposed that would improve the processing speed by either acceleration in hardware or algorithmic changes.

ITC/USA 2010 Conference Proceedings / The Forty-Sixth Annual International Telemetering Conference and Technical Exhibition / October 25-28, 2010 / Town and Country Resort & Convention Center, San Diego, California
Synchronized video is crucial for data acquisition and telecommunication applications. For real-time applications, out-of-sync video may cause jitter, choppiness and latency. For data analysis, it is important to synchronize multiple video channels and data that are acquired from PCM, MIL-STD-1553 and other sources. Nowadays, video codecs can be easily obtained to play most types of video. However, a great deal of effort is still required to develop the synchronization methods that are used in a data acquisition system. This paper will describe several methods that TTC has adopted in our system to improve the synchronization of multiple data sources.

The buzz surrounding artificial intelligences continues to grow. They are currently used in a wide variety of systems and appliances, such as video games, virtual personal assistants, and self-driving cars. This paper explores the possibility of a self-learning AI that can play the classic arcade game Q*BERT, using only screenshots as input. It is tested to work on several different screens sizes, and the results are collected and compared to that of a human player, as well as results from previous research. The results are fairly positive. While the AI had a hard time of matching the human player on average score, it did get close to the highest score.

This project extended COLLDESIGN, an interactive collaborative modeling tool that was developed by Mr. Hara Totapally. The initial version included a collaborative framework comprised of configurable client and server components. This project accomplished a complete implementation of the Class Diagram view. In addition, extending the framework, text messaging and audio conferencing features have been implemented to allow for real-time textual and audio communication between team members working on a particular project. VideoClient is the GUI of the application.

Real-time conferencing and collaborative computing is a great way to make developers more effective. This project is a collaborative framework development comprising configurable client and server components.

Thesis (M. S.)--Electrical and Computer Engineering, Georgia Institute of Technology, 2009.
Committee Chair: Jayant, Nikil; Committee Member: Altunbasak, Yucel; Committee Member: Sivakumar, Raghupathy. Part of the SMARTech Electronic Thesis and Dissertation Collection.

Práce představuje novou adaptabilní metodu pro on-line vyhledávání videa v reálném čase pomocí vizuálních slovníků. Nová metoda se zaměřuje na nízkou výpočetní náročnost a přesnost vyhledání při on-line použití. Metoda vychází z technik využitých u statických vizuálních slovníků. Tyto běžné techniky jsou upraveny tak, aby byly schopné se adaptovat na proměnlivá data. Postupy, které toto u nové metody řeší, jsou - dynamická inverzní frekvence dokumentů, adaptabilní vizuální slovník a proměnlivý invertovaný index. Navržený postup byl vyhodnocen na úloze vyhledávání videa a prezentované výsledky ukazují, jaké vlastnosti má adaptabilní metoda ve srovnání se statickým přístupem. Nová adaptabilní metoda je založena na konceptu plovoucího okna, který definuje, jakým způsobem se vybírají data pro adaptaci a ke zpracování. Společně s konceptem je definován i matematický aparát, který umožňuje vyhodnotit, jak koncept nejlépe využít pro různé metody zpracování videa. Praktické využití adaptabilní metody je konkrétně u systémů pro zpracování videa, kde se očekává změna v charakteru vizuálních dat nebo tam, kde není předem známo, jakého charakteru vizuální data budou.

Nous étudions dans cette thèse la possibilité de reconnaitre des objets dans des images compressées, sans les reconstruire. L'algorithme de compression le plus adapte semble celui fonde sur l'extraction des contours multi échelle quinconce des images. Le problème de la reconnaissance nous amène à introduire un nouvel outil de comparaison d'images binaires: la distance de Hausdorff censurée. Cet outil s'est avéré robuste et rapide à calculer. Ces deux points sont étudiés avec soin. Cette distance est enfin utilisée pour reconnaitre et localiser des objets spécifiques dans des scènes de grande taille. Nous proposons trois approches multi échelles pour résoudre ce problème, qui prennent en compte le fait que l'objet recherché peut être en partie caché, ou qu'il peut être vu sous un angle différent de son modèle. L'algorithme que nous avons développé est rapide sur station de travail classique. Sa robustesse a été étudiée soigneusement. Sa parallélisation nous permet d'atteindre le temps réel dans un cadre opérationnel raisonnable

Tang Wai-Kwan.
Thesis (M.Phil.)--Chinese University of Hong Kong, 2001.
Includes bibliographical references (leaves 88-95).
Abstracts in English and Chinese.
Abstract --- p.i
Acknowledgement --- p.iii
Chapter 1 --- Introduction --- p.1
Chapter 1.1 --- Applications --- p.3
Chapter 1.2 --- Motivation and Goal --- p.6
Chapter 1.3 --- Thesis Outline --- p.7
Chapter 2 --- Background and Related Work --- p.8
Chapter 2.1 --- Panoramic Image Navigation --- p.8
Chapter 2.2 --- Image Mosaicing --- p.9
Chapter 2.2.1 --- Image Registration --- p.10
Chapter 2.2.2 --- Image Composition --- p.12
Chapter 2.3 --- Immersive Display --- p.13
Chapter 2.4 --- Video Streaming --- p.14
Chapter 2.4.1 --- Video Coding --- p.15
Chapter 2.4.2 --- Transport Protocol --- p.18
Chapter 3 --- System Design --- p.19
Chapter 3.1 --- System Architecture --- p.19
Chapter 3.1.1 --- Video Capture Module --- p.19
Chapter 3.1.2 --- Video Streaming Module --- p.23
Chapter 3.1.3 --- Stitching and Rendering Module --- p.24
Chapter 3.1.4 --- Display Module --- p.24
Chapter 3.2 --- Design Issues --- p.25
Chapter 3.2.1 --- Modular Design --- p.25
Chapter 3.2.2 --- Scalability --- p.26
Chapter 3.2.3 --- Workload distribution --- p.26
Chapter 4 --- Panoramic Video Mosaic --- p.28
Chapter 4.1 --- Video Mosaic to Image Mosaic --- p.28
Chapter 4.1.1 --- Assumptions --- p.29
Chapter 4.1.2 --- Processing Pipeline --- p.30
Chapter 4.2 --- Camera Calibration --- p.33
Chapter 4.2.1 --- Perspective Projection --- p.33
Chapter 4.2.2 --- Distortion --- p.36
Chapter 4.2.3 --- Calibration Procedure --- p.37
Chapter 4.3 --- Panorama Generation --- p.39
Chapter 4.3.1 --- Cylindrical and Spherical Panoramas --- p.39
Chapter 4.3.2 --- Homography --- p.41
Chapter 4.3.3 --- Homography Computation --- p.42
Chapter 4.3.4 --- Error Minimization --- p.44
Chapter 4.3.5 --- Stitching Multiple Images --- p.46
Chapter 4.3.6 --- Seamless Composition --- p.47
Chapter 4.4 --- Image Mosaic to Video Mosaic --- p.49
Chapter 4.4.1 --- Varying Intensity --- p.49
Chapter 4.4.2 --- Video Frame Management --- p.50
Chapter 5 --- Immersive Display --- p.52
Chapter 5.1 --- Human Perception System --- p.52
Chapter 5.2 --- Creating Virtual Scene --- p.53
Chapter 5.3 --- VisionStation --- p.54
Chapter 5.3.1 --- F-Theta Lens --- p.55
Chapter 5.3.2 --- VisionStation Geometry --- p.56
Chapter 5.3.3 --- Sweet Spot Relocation and Projection --- p.57
Chapter 5.3.4 --- Sweet Spot Relocation in Vector Representation --- p.61
Chapter 6 --- Video Streaming --- p.65
Chapter 6.1 --- Video Compression --- p.66
Chapter 6.2 --- Transport Protocol --- p.66
Chapter 6.3 --- Latency and Jitter Control --- p.67
Chapter 6.4 --- Synchronization --- p.70
Chapter 7 --- Implementation and Results --- p.71
Chapter 7.1 --- Video Capture --- p.71
Chapter 7.2 --- Video Streaming --- p.73
Chapter 7.2.1 --- Video Encoding --- p.73
Chapter 7.2.2 --- Streaming Protocol --- p.75
Chapter 7.3 --- Implementation Results --- p.76
Chapter 7.3.1 --- Indoor Scene --- p.76
Chapter 7.3.2 --- Outdoor Scene --- p.78
Chapter 7.4 --- Evaluation --- p.78
Chapter 8 --- Conclusion --- p.83
Chapter 8.1 --- Summary --- p.83
Chapter 8.2 --- Future Directions --- p.84
Chapter A --- Parallax --- p.86

by Walter Chi-Woon Fung.
Thesis (M.Phil.)--Chinese University of Hong Kong, 1998.
Includes bibliographical references (leaves 64-[67]).
Abstract also in Chinese.
Chapter 1 --- Introduction --- p.1
Chapter 1.1 --- Background --- p.1
Chapter 1.2 --- Limitation of Existing Research --- p.3
Chapter 1.3 --- Contributions of This Thesis --- p.3
Chapter 1.4 --- Organization of the Thesis --- p.4
Chapter 2 --- Related Work --- p.5
Chapter 2.1 --- Ongoing Efforts For The Support of Real Time Applications on the Internet - RTP --- p.5
Chapter 2.2 --- Using the Algorithm on top of RTP --- p.7
Chapter 3 --- An Adaptive Video Retrieval Algorithm --- p.9
Chapter 3.1 --- Lossless Environment --- p.9
Chapter 3.1.1 --- Adapting the Request Rate to the Available Bandwidth --- p.12
Chapter 3.2 --- Lossy Environment --- p.17
Chapter 3.2.1 --- Adapting Ar in Lossy Environment --- p.20
Chapter 3.3 --- Adjusting the Window Size --- p.24
Chapter 3.4 --- Measurement Issues --- p.27
Chapter 3.5 --- Mapping between Data Rate and Frame Rate --- p.28
Chapter 4 --- Rate Measurement --- p.30
Chapter 4.1 --- Arrival Rate Estimation --- p.30
Chapter 4.2 --- Loss Rate Estimation --- p.32
Chapter 5 --- Frame Skipping and Stuffing --- p.37
Chapter 5.1 --- MPEG-1 Video Stream Basics --- p.37
Chapter 5.2 --- Frame Skipping --- p.38
Chapter 5.3 --- Frame Stuffing In Lossy Environment --- p.40
Chapter 6 --- Experiment Result and Analysis --- p.43
Chapter 6.1 --- Experiment --- p.43
Chapter 6.2 --- Analysis --- p.54
Chapter 6.2.1 --- Interacting With Streams With No Rate Control --- p.56
Chapter 6.2.2 --- Multiple Streams Running The Algorithm --- p.58
Chapter 6.2.3 --- Calculation of p --- p.59
Chapter 7 --- Conclusions --- p.61
Bibliography --- p.64