In recent years Content Based Information Retrieval (CBIR) has received a lot of research attention, starting with audio, followed by images and video. Video ngerprinting is a CBIR technique that creates a digital descriptor, also known as a ngerprint, for videos based on its content. These ngerprints are then saved to a database and used to detect unknown videos by comparing the unknown video's ngerprint to the ngerprints in the database to get a match. Many techniques have already been proposed with various levels of success, but most of the existing techniques focus mainly on robustness and neglect the speed of implementation. In this dissertation a novel video ngerprinting technique will be developed with the main focus on detecting advertisements in a television broadcast. Therefore the system must be able to process the incoming video stream in real-time and detect all the advertisements that are present. Even though the algorithm has to be fast, it still has to be robust enough to handle a moderate amount of distortions. These days video ngerprinting still holds many challenges as it involves characterizing videos, made up of sequences of images, e ectively. This means the algorithm must somehow imitate the inherent ability of humans to recognize a video almost instantly. The technique uses the content of the video to derive a ngerprint, thus the features used by the ngerprinting algorithm should be robust to distortions that don't a ect content according to humans. / Thesis (MIng (Computer and Electronic Engineering))--North-West University, Potchefstroom Campus, 2013

Today, the advent of networking technologies and computer hardware have enabled more and more inexpensive PCs, various mobile devices, smart phones, PDAs, sensors and cameras to be linked to the Internet with better connectivity. In recent years, we have witnessed the emergence of several instances of distributed applications, providing infrastructures for social interactions over large-scale wide-area networks and facilitating the ways users share and publish data. User generated data today range from simple text files to (semi-) structured documents and multimedia content. With the emergence of Semantic Web, the number of features (associated with a content) that are used in order to index those large amounts of heterogenous pieces of data is growing dramatically. The feature sets associated with each content type can grow continuously as we discover new ways of describing a content in formulated terms. As the number of dimensions in the feature data grow (as high as 100 to 1000), it becomes harder and harder to search for information in a dataset due to the curse of dimensionality and it is not appropriate to use naive search methods, as their performance degrade to linear search. As an alternative, we can distribute the content and the query processing load to a set of peers in a distributed Peer-to-Peer (P2P) network and incorporate high-dimensional distributed search techniques to attack the problem. Currently, a large percentage of Internet traffic consists of video and music files shared and exchanged over P2P networks. In most present services, searching for music is performed through keyword search and naive string-matching algorithms using collaborative filtering techniques which mostly use tag based approaches. In music information retrieval (MIR) systems, the main goal is to make recommendations similar to the music that the user listens to. In these systems, techniques based on acoustic feature extraction can be employed to achieve content-based music similarity search (i.e., searching through music based on what can be heard from the music track). Using these techniques we can devise an automated measure of similarity that can replace the need for human experts (or users) who assign descriptive genre tags and meta-data to each recording and solve the famous cold-start problem associated with the collaborative filtering techniques. In this work we explore the advantages of distributed structures by efficiently distributing the content features and query processing load on the peers in a P2P network. Using a family of Locality Sensitive Hash (LSH) functions based on p-stable distributions we propose an efficient, scalable and load-balanced system, capable of performing K-Nearest-Neighbor (KNN) and Range queries. We also propose a new load-balanced indexing algorithm and evaluate it using our Java based simulator. Our results show that this P2P design ensures load-balancing and guarantees logarithmic number of hops for query processing. Our system is extensible to be used with all types of multi-dimensional feature data and it can also be employed as the main indexing scheme of a multipurpose recommendation system. / Graduate

Thesis (Ph.D.)--University of Wollongong, 2003. / Typescript. Includes bibliographical references: leaf 191-198.

Recent advances in technology have increased awareness of the necessity for automated systems in people’s everyday lives. Artificial systems are more frequently being introduced into environments previously thought to be too perilous for humans to operate in. Some robots can be used to extract potentially hazardous materials from sites inaccessible to humans, while others are being developed to aid humans with laborious tasks. A crucial aspect of all artificial systems is the manner in which they interact with their immediate surroundings. Developing such a deceivingly simply aspect has proven to be significantly challenging, as it not only entails the methods through which the system perceives its environment, but also its ability to perform critical tasks. These undertakings often involve the coordination of numerous subsystems, each performing its own complex duty. To complicate matters further, it is nowadays becoming increasingly important for these artificial systems to be able to perform their tasks in real-time. The task of object recognition is typically described as the process of retrieving the object in a database that is most similar to an unknown, or query, object. Pose estimation, on the other hand, involves estimating the position and orientation of an object in three-dimensional space, as seen from an observer’s viewpoint. These two tasks are regarded as vital to many computer vision techniques and and regularly serve as input to more complex perception algorithms. An approach is presented which regards the object recognition and pose estimation procedures as mutually dependent. The core idea is that dissimilar objects might appear similar when observed from certain viewpoints. A feature-based conceptualisation, which makes use of a database, is implemented and used to perform simultaneous object recognition and pose estimation. The design incorporates data compression techniques, originally suggested by the image-processing community, to facilitate fast processing of large databases. System performance is quantified primarily on object recognition, pose estimation and execution time characteristics. These aspects are investigated under ideal conditions by exploiting three-dimensional models of relevant objects. The performance of the system is also analysed for practical scenarios by acquiring input data from a structured light implementation, which resembles that obtained from many commercial range scanners. Practical experiments indicate that the system was capable of performing simultaneous object recognition and pose estimation in approximately 230 ms once a novel object has been sensed. An average object recognition accuracy of approximately 73% was achieved. The pose estimation results were reasonable but prompted further research. The results are comparable to what has been achieved using other suggested approaches such as Viewpoint Feature Histograms and Spin Images. / Dissertation (MEng)--University of Pretoria, 2013. / gm2014 / Electrical, Electronic and Computer Engineering / unrestricted

Les besoins constants d’innovation en matière de performances et d’économie des ressources nous poussent à effectuer des optimisations dans la conception et l’utilisation des outils cryptographiques. Cela nous amène à étudier plusieurs aspects dans cette thèse : les algorithmes cryptographiques parallèles, les algorithmes cryptographiques incrémentaux et les dictionnaires authentifiés.Dans le cadre de la cryptographie parallèle, nous nous intéressons aux fonctions de hachage basées sur des arbres. Nous montrons en particulier quelles structures arborescentes utiliser pour atteindre un temps d’exécution optimum avec un nombre de processeurs que nous cherchons à minimiser dans un second temps. Nous étudions également d'autres formesd'arborescence favorisant l'équité et la scalabilité.Les systèmes cryptographiques incrémentaux permettent, lorsque nous modifions des documents, de mettre à jour leurs formes cryptographiques efficacement. Nous montrons que les systèmes actuels restreignent beaucoup trop les modifications possibles et introduisons de nouveaux algorithmes s’appuyant sur ces derniers, utilisés comme des boites noires, afin de rendre possible une large gamme de modifications aux documents tout en conservant une propriété de secret de l’opération effectuée.Notre intérêt porte ensuite sur les dictionnaires authentifiés, utilisés pour authentifier les réponses aux requêtes des utilisateurs sur un dictionnaire, en leur fournissant une preuve d’authenticité pour chaque réponse. Nous nous focalisons sur des systèmes basés sur des arbres de hachage et proposons une solution pour amoindrir leur principal inconvénient, celui de la taille des preuves. / The need for continuing innovation in terms of performances and resource savings impel us to optimize the design and the use of cryptographic mechanisms. This leads us to consider several aspects in this dissertation: parallel cryptographic algorithms, incremental cryptographic algorithms and authenticated dictionaries.In the context of parallel cryptography we are interested in hash functions. In particular, we show which tree structures to use to reach an optimal running time. For this running time, we show how to decrease the amount of involved processors. We also explore alternative (sub-optimal) tree structures which decrease the number of synchronizations in multithreaded implementations while balancing at best the load of the work among the threads.Incremental cryptographic schemes allow the efficient updating of cryptographic forms when we change some blocks of the corresponding documents. We show that the existing incremental schemes restrict too much the possible modification operations. We then introduce new algorithms which use these ones as black boxes to allow a broad range of modification operations, while preserving a privacy property about these operations.We then turn our attention to authenticated dictionaries which are used to authenticate answers to queries on a dictionary, by providing to users an authentication proof for each answer. We focus on authenticated dictionaries based on hash trees and we propose a solution to remedy their main shortcoming, the size of proofs provided to users.

Recent advances in technology have increased awareness of the necessity for automated systems in people’s everyday lives. Artificial systems are more frequently being introduced into environments previously thought to be too perilous for humans to operate in. Some robots can be used to extract potentially hazardous materials from sites inaccessible to humans, while others are being developed to aid humans with laborious tasks. A crucial aspect of all artificial systems is the manner in which they interact with their immediate surroundings. Developing such a deceivingly simply aspect has proven to be significantly challenging, as it not only entails the methods through which the system perceives its environment, but also its ability to perform critical tasks. These undertakings often involve the coordination of numerous subsystems, each performing its own complex duty. To complicate matters further, it is nowadays becoming increasingly important for these artificial systems to be able to perform their tasks in real-time. The task of object recognition is typically described as the process of retrieving the object in a database that is most similar to an unknown, or query, object. Pose estimation, on the other hand, involves estimating the position and orientation of an object in three-dimensional space, as seen from an observer’s viewpoint. These two tasks are regarded as vital to many computer vision techniques and regularly serve as input to more complex perception algorithms. An approach is presented which regards the object recognition and pose estimation procedures as mutually dependent. The core idea is that dissimilar objects might appear similar when observed from certain viewpoints. A feature-based conceptualisation, which makes use of a database, is implemented and used to perform simultaneous object recognition and pose estimation. The design incorporates data compression techniques, originally suggested by the image-processing community, to facilitate fast processing of large databases. System performance is quantified primarily on object recognition, pose estimation and execution time characteristics. These aspects are investigated under ideal conditions by exploiting three-dimensional models of relevant objects. The performance of the system is also analysed for practical scenarios by acquiring input data from a structured light implementation, which resembles that obtained from many commercial range scanners. Practical experiments indicate that the system was capable of performing simultaneous object recognition and pose estimation in approximately 230 ms once a novel object has been sensed. An average object recognition accuracy of approximately 73% was achieved. The pose estimation results were reasonable but prompted further research. The results are comparable to what has been achieved using other suggested approaches such as Viewpoint Feature Histograms and Spin Images. / Dissertation (MEng)--University of Pretoria, 2013. / gm2014 / Electrical, Electronic and Computer Engineering / unrestricted

Cette thèse étudie deux tâches fondamentales d'apprentissage non supervisé: la recherche des plus proches voisins et le clustering de données massives en grande dimension pour respecter d'importantes contraintes de temps et d'espace.Tout d'abord, un nouveau cadre théorique permet de réduire le coût spatial et d'augmenter le débit de traitement du Cross-polytope LSH pour la recherche du plus proche voisin presque sans aucune perte de précision.Ensuite, une méthode est conçue pour apprendre en une seule passe sur des données en grande dimension des codes compacts binaires. En plus de garanties théoriques, la qualité des sketches obtenus est mesurée dans le cadre de la recherche approximative des plus proches voisins. Puis, un algorithme de clustering sans paramètre et efficace en terme de coût de stockage est développé en s'appuyant sur l'extraction d'un arbre couvrant minimum approché du graphe de dissimilarité compressé auquel des coupes bien choisies sont effectuées. / This thesis focuses on how to perform efficiently unsupervised machine learning such as the fundamentally linked nearest neighbor search and clustering task, under time and space constraints for high-dimensional datasets. First, a new theoretical framework reduces the space cost and increases the rate of flow of data-independent Cross-polytope LSH for the approximative nearest neighbor search with almost no loss of accuracy.Second, a novel streaming data-dependent method is designed to learn compact binary codes from high-dimensional data points in only one pass. Besides some theoretical guarantees, the quality of the obtained embeddings are accessed on the approximate nearest neighbors search task.Finally, a space-efficient parameter-free clustering algorithm is conceived, based on the recovery of an approximate Minimum Spanning Tree of the sketched data dissimilarity graph on which suitable cuts are performed.

Nous avons précédemment mené une étude comparative entre les descripteurs FV et CNN dans le cadre de la recherche par similarité d’instance. Cette étude montre notamment que les descripteurs issus de CNN manquent d’invariance aux transformations comme les rotations ou changements d’échelle. Nous montrons dans un premier temps comment des réductions de dimension (“pooling”) appliquées sur la base de données d’images permettent de réduire fortement l’impact de ces problèmes. Certaines variantes préservent la dimensionnalité des descripteurs associés à une image, alors que d’autres l’augmentent, au prix du temps d’exécution des requêtes. Dans un second temps, nous proposons la réduction de dimension emboitée pour l’invariance (NIP), une méthode originale pour la production, à partir de descripteurs issus de CNN, de descripteurs globaux invariants à de multiples transformations. La méthode NIP est inspirée de la théorie pour l’invariance “i-theory”, une théorie mathématique proposée il y a peu pour le calcul de transformations invariantes à des groupes au sein de réseaux de neurones acycliques. Nous montrons que NIP permet d’obtenir des descripteurs globaux compacts (mais non binaires) et robustes aux rotations et aux changements d’échelle, que NIP est plus performants que les autres méthodes à dimensionnalité équivalente sur la plupart des bases de données d’images. Enfin, nous montrons que la combinaison de NIP avec la méthode de hachage RBMH proposée précédemment permet de produire des codes binaires à la fois compacts et invariants à plusieurs types de transformations. La méthode NIP+RBMH, évaluée sur des bases de données d’images de moyennes et grandes échelles, se révèle plus performante que l’état de l’art, en particulier dans le cas de descripteurs binaires de très petite taille (de 32 à 256 bits). / Image instance retrieval is the problem of finding an object instance present in a query image from a database of images. Also referred to as particular object retrieval, this problem typically entails determining with high precision whether the retrieved image contains the same object as the query image. Scale, rotation and orientation changes between query and database objects and background clutter pose significant challenges for this problem. State-of-the-art image instance retrieval pipelines consist of two major steps: first, a subset of images similar to the query are retrieved from the database, and second, Geometric Consistency Checks (GCC) are applied to select the relevant images from the subset with high precision. The first step is based on comparison of global image descriptors: high-dimensional vectors with up to tens of thousands of dimensions rep- resenting the image data. The second step is computationally highly complex and can only be applied to hundreds or thousands of images in practical applications. More discriminative global descriptors result in relevant images being more highly ranked, resulting in fewer images that need to be compared pairwise with GCC. As a result, better global descriptors are key to improving retrieval performance and have been the object of much recent interest. Furthermore, fast searches in large databases of millions or even billions of images requires the global descriptors to be compressed into compact representations. This thesis will focus on how to achieve extremely compact global descriptor representations for large-scale image instance retrieval. After introducing background concepts about supervised neural networks, Restricted Boltzmann Machine (RBM) and deep learning in Chapter 2, Chapter 3 will present the design principles and recent work for the Convolutional Neural Networks (CNN), which recently became the method of choice for large-scale image classification tasks. Next, an original multistage approach for the fusion of the output of multiple CNN is proposed. Submitted as part of the ILSVRC 2014 challenge, results show that this approach can significantly improve classification results. The promising perfor- mance of CNN is largely due to their capability to learn appropriate high-level visual representations from the data. Inspired by a stream of recent works showing that the representations learnt on one particular classification task can transfer well to other classification tasks, subsequent chapters will focus on the transferability of representa- tions learnt by CNN to image instance retrieval…

Energy conservation is a major concern in todays data centers, which are the 21st century data processing factories, and where large and complex software systems such as distributed data management stores run and serve billions of users. The two main drivers of this major concern are the pollution impact data centers have on the environment due to their waste heat, and the expensive cost data centers incur due to their enormous energy demand. Among the many subsystems of data centers, the storage system is one of the main sources of energy consumption. Among the many types of storage systems, key/value stores happen to be the widely used in the data centers. In this work, I investigate energy saving techniques that enable a consistent hash based key/value store save energy during low activity times, and whenever there is an opportunity to reuse the waste heat of data centers.

Energy conservation is a major concern in todays data centers, which are the 21st century data processing factories, and where large and complex software systems such as distributed data management stores run and serve billions of users. The two main drivers of this major concern are the pollution impact data centers have on the environment due to their waste heat, and the expensive cost data centers incur due to their enormous energy demand. Among the many subsystems of data centers, the storage system is one of the main sources of energy consumption. Among the many types of storage systems, key/value stores happen to be the widely used in the data centers. In this work, I investigate energy saving techniques that enable a consistent hash based key/value store save energy during low activity times, and whenever there is an opportunity to reuse the waste heat of data centers.