This thesis addresses the task of developing automatic algorithms for analysing the two-dimensional ultrasound video footage obtained from fetal heart screening scans. These scans are typically performed in the second trimester of pregnancy to check for congenital heart anomalies and require significant training and anatomical knowledge to perform. The aim is to develop a tool that runs at high frame rates with no user initialisation and infers the visibility, position, orientation, view classification, and cardiac phase of the heart, and additionally the locations of cardiac structures of interest (such as valves and vessels) in a manner that is robust to the various sources of variation that occur in real-world ultrasound scanning. This is the first work to attempt such a detailed automated analysis of these videos. The problem is posed as a Bayesian filtering problem, which provides a principled framework for aggregating uncertain measurements across a number of frames whilst exploiting the constraints imposed by anatomical feasibility. The resulting inference problem is solved approximately with a particle filter, whose state space is partitioned to reduce the problems associated with filtering in high-dimensional spaces. Rotation-invariant features are captured from the videos in an efficient way in order to tackle the problem of unknown orientation. These are used within random forest learning models, including a novel formulation to predict circular-valued variables. The algorithm is validated on an annotated clinical dataset, and the results are compared to estimates of inter- and intra-observer variation, which are significant in both cases due to the inherent ambiguity in the imagery. The results suggest that the algorithm's output approaches these benchmarks in several respects, and fall slightly behind in others. The work presented here is an important first step towards developing automated clinical tools for the detection of congenital heart disease.

L'Interprétation automatique de vidéos est un horizon qui demeure difficile a atteindre en utilisant les approches actuelles de vision par ordinateur. Une des principales difficultés est d'aller au-delà des descripteurs visuels actuels (de même que pour les autres modalités, audio, textuelle, etc) pour pouvoir mettre en oeuvre des algorithmes qui permettraient de reconnaitre automatiquement des sections de vidéos, potentiellement longues, dont le contenu appartient à une certaine catégorie définie de manière sémantique. Un exemple d'une telle section de vidéo serait une séquence ou une personne serait en train de pêcher; un autre exemple serait une dispute entre le héros et le méchant dans un film d'action hollywoodien. Dans ce manuscrit, nous présentons plusieurs contributions qui vont dans le sens de cet objectif ambitieux, en nous concentrant sur trois tâches d'analyse de vidéos: le résumé automatique, la classification, la localisation temporelle.Tout d'abord, nous introduisons une approche pour le résumé automatique de vidéos, qui fournit un résumé de courte durée et informatif de vidéos pouvant être très longues, résumé qui est de plus adapté à la catégorie de vidéos considérée. Nous introduisons également une nouvelle base de vidéos pour l'évaluation de méthodes de résumé automatique, appelé MED-Summaries, ou chaque plan est annoté avec un score d'importance, ainsi qu'un ensemble de programmes informatiques pour le calcul des métriques d'évaluation.Deuxièmement, nous introduisons une nouvelle base de films de cinéma annotés, appelée Inria Action Movies, constitué de films d'action hollywoodiens, dont les plans sont annotés suivant des catégories sémantiques non-exclusives, dont la définition est suffisamment large pour couvrir l'ensemble du film. Un exemple de catégorie est "course-poursuite"; un autre exemple est "scène sentimentale". Nous proposons une approche pour localiser les sections de vidéos appartenant à chaque catégorie et apprendre les dépendances temporelles entre les occurrences de chaque catégorie.Troisièmement, nous décrivons les différentes versions du système développé pour la compétition de détection d'événement vidéo TRECVID Multimédia Event Detection, entre 2011 et 2014, en soulignant les composantes du système dont l'auteur du manuscrit était responsable. / Automatic interpretation and understanding of videos still remains at the frontier of computer vision. The core challenge is to lift the expressive power of the current visual features (as well as features from other modalities, such as audio or text) to be able to automatically recognize typical video sections, with low temporal saliency yet high semantic expression. Examples of such long events include video sections where someone is fishing (TRECVID Multimedia Event Detection), or where the hero argues with a villain in a Hollywood action movie (Inria Action Movies). In this manuscript, we present several contributions towards this goal, focusing on three video analysis tasks: summarization, classification, localisation.First, we propose an automatic video summarization method, yielding a short and highly informative video summary of potentially long videos, tailored for specified categories of videos. We also introduce a new dataset for evaluation of video summarization methods, called MED-Summaries, which contains complete importance-scorings annotations of the videos, along with a complete set of evaluation tools.Second, we introduce a new dataset, called Inria Action Movies, consisting of long movies, and annotated with non-exclusive semantic categories (called beat-categories), whose definition is broad enough to cover most of the movie footage. Categories such as "pursuit" or "romance" in action movies are examples of beat-categories. We propose an approach for localizing beat-events based on classifying shots into beat-categories and learning the temporal constraints between shots.Third, we overview the Inria event classification system developed within the TRECVID Multimedia Event Detection competition and highlight the contributions made during the work on this thesis from 2011 to 2014.

Avec la récente et importante croissance des contenus vidéos, la compréhension automatique de vidéos est devenue un problème majeur.Ce mémoire présente plusieurs contributions sur deux tâches de la compréhension automatique de vidéos : l'estimation du flot optique et la localisation d'actions humaines.L'estimation du flot optique consiste à calculer le déplacement de chaque pixel d'une vidéo et fait face à plusieurs défis tels que les grands déplacements non rigides, les occlusions et les discontinuités du mouvement.Nous proposons tout d'abord une méthode pour le calcul du flot optique, basée sur un modèle variationnel qui incorpore une nouvelle méthode d'appariement.L'algorithme d'appariement proposé repose sur une architecture corrélationnelle hiérarchique à plusieurs niveaux et gère les déformations non rigides ainsi que les textures répétitives.Il permet d'améliorer l'estimation du flot en présence de changements d'apparence significatifs et de grands déplacements.Nous présentons également une nouvelle approche pour l'estimation du flot optique basée sur une interpolation dense de correspondances clairsemées tout en respectant les contours.Cette méthode tire profit d'une distance géodésique basée sur les contours qui permet de respecter les discontinuités du mouvement et de gérer les occlusions.En outre, nous proposons une approche d'apprentissage pour détecter les discontinuités du mouvement.Les motifs de discontinuité du mouvement sont prédits au niveau d'un patch en utilisant des forêts aléatoires structurées.Nous montrons expérimentalement que notre approche surclasse la méthode basique construite sur le gradient du flot tant sur des données synthétiques que sur des vidéos réelles.Nous présentons à cet effet une base de données contenant des vidéos d'utilisateurs.La localisation d'actions humaines consiste à reconnaître les actions présentes dans une vidéo, comme `boire' ou `téléphoner', ainsi que leur étendue temporelle et spatiale.Nous proposons tout d'abord une nouvelle approche basée sur les réseaux de neurones convolutionnels profonds.La méthode passe par l'extraction de tubes dépendants de la classe à détecter, tirant parti des dernières avancées en matière de détection et de suivi.La description des tubes est enrichie par des descripteurs spatio-temporels locaux.La détection temporelle est effectuée à l'aide d'une fenêtre glissante à l'intérieur de chaque tube.Notre approche surclasse l'état de l'art sur des bases de données difficiles de localisation d'actions.Deuxièmement, nous présentons une méthode de localisation d'actions faiblement supervisée, c'est-à-dire qui ne nécessite pas l'annotation de boîtes englobantes.Des candidats de localisation d'actions sont calculés en extrayant des tubes autour des humains.Cela est fait en utilisant un détecteur d'humains robuste aux poses inhabituelles et aux occlusions, appris sur une base de données de poses humaines.Un rappel élevé est atteint avec seulement quelques tubes, permettant d'appliquer un apprentissage à plusieurs instances.En outre, nous présentons une nouvelle base de données pour la localisation d'actions humaines.Elle surmonte les limitations des bases existantes, telles la diversité et la durée des vidéos.Notre approche faiblement supervisée obtient des résultats proches de celles totalement supervisées alors qu'elle réduit significativement l'effort d'annotations requis. / With the recent overwhelming growth of digital video content, automatic video understanding has become an increasingly important issue.This thesis introduces several contributions on two automatic video understanding tasks: optical flow estimation and human action localization.Optical flow estimation consists in computing the displacement of every pixel in a video andfaces several challenges including large non-rigid displacements, occlusions and motion boundaries.We first introduce an optical flow approach based on a variational model that incorporates a new matching method.The proposed matching algorithm is built upon a hierarchical multi-layer correlational architecture and effectively handles non-rigid deformations and repetitive textures.It improves the flow estimation in the presence of significant appearance changes and large displacements.We also introduce a novel scheme for estimating optical flow based on a sparse-to-dense interpolation of matches while respecting edges.This method leverages an edge-aware geodesic distance tailored to respect motion boundaries and to handle occlusions.Furthermore, we propose a learning-based approach for detecting motion boundaries.Motion boundary patterns are predicted at the patch level using structured random forests.We experimentally show that our approach outperforms the flow gradient baseline on both synthetic data and real-world videos,including an introduced dataset with consumer videos.Human action localization consists in recognizing the actions that occur in a video, such as `drinking' or `phoning', as well as their temporal and spatial extent.We first propose a novel approach based on Deep Convolutional Neural Network.The method extracts class-specific tubes leveraging recent advances in detection and tracking.Tube description is enhanced by spatio-temporal local features.Temporal detection is performed using a sliding window scheme inside each tube.Our approach outperforms the state of the art on challenging action localization benchmarks.Second, we introduce a weakly-supervised action localization method, ie, which does not require bounding box annotation.Action proposals are computed by extracting tubes around the humans.This is performed using a human detector robust to unusual poses and occlusions, which is learned on a human pose benchmark.A high recall is reached with only several human tubes, allowing to effectively apply Multiple Instance Learning.Furthermore, we introduce a new dataset for human action localization.It overcomes the limitations of existing benchmarks, such as the diversity and the duration of the videos.Our weakly-supervised approach obtains results close to fully-supervised ones while significantly reducing the required amount of annotations.

Pour de nombreux pays européens, ces dernières années ont été marquées par l’introduction des Enseignements Scientifiques Fondés sur l’Investigation (ESFI) dans les programmes scolaires. Deux objectifs sont poursuivis : offrir une image « plus conforme » de l’activité scientifique et éveiller l’intérêt des élèves en mettant l’accent sur leur rôle actif dans le processus d’apprentissage. Parallèlement à l’introduction de ces démarches, la mise en œuvre d’évaluations formatives ouvre une voie prometteuse pour répondre à ces objectifs. Elles sont en effet susceptibles de développer l’autorégulation des apprentissages des élèves et de constituer une aide pertinente pour les enseignants afin de réguler leur enseignement (Allal & Mottier Lopez, 2007 ; Clark, 2012 ; Wiliam, 2010). Cette recherche doctorale traite des impacts de différentes modalités d’évaluation formative sur l’autorégulation des apprentissages des élèves dans le cadre spécifique des ESFI. Deux enjeux ont guidé ce travail. Le premier, empirique, concernait l’analyse des pratiques évaluatives des enseignants et de leurs effets sur les processus d’autorégulation. Le second visait la mise au point d’une méthodologie qui permette d’analyser les variables étudiées. Pour ce faire, nous avons procédé par enregistrements vidéo de séances de classe et construit des grilles d’indicateurs nous permettant d’analyser les processus in situ. Différentes situations d’enseignement ont été comparées. La première correspondait aux évaluations formatives telles que les enseignants la mettent en œuvre dans leurs pratiques quotidiennes. La deuxième concernait les pratiques évaluatives mises en œuvre par les mêmes enseignants l’année suivante, après qu’un retour réflexif sur leur séance ait été conduit. Nous montrons un meilleur équilibre dans l’usage des différentes modalités d’évaluation formative dans la deuxième situation, notamment vers une plus grande responsabilisation des élèves et mise en avant des pairs en tant que ressource. Les élèves font aussi preuve d’une autorégulation de leur comportement plus efficace au regard d’un temps passé à produire des stratégies de résolution plus important et d’un meilleur engagement dans la tâche. Des manques sont néanmoins relevés quant à la façon d’intégrer formellement l’outil d’autoévaluation à l’activité des élèves. Des pistes d’action pour combiner efficacement l’évaluation aux différentes tâches d’apprentissage sont alors proposées. / Since a few years, in many European countries, Inquiry-Based Science Education (IBSE) has impacted science curriculums. Two goals are at stake: to provide an image of scientific activity more consistent with the actual activity of scientists, and to arouse students' interest by emphasizing their active role in the learning process. With the introduction of these measures, the implementation of formative assessments opens a promising way to meet these goals. They are in fact likely to develop students’ self-regulation and to provide relevant feedbacks for teachers to regulate their teaching (Allal & Mottier Lopez, 2007; Clark, 2012; Wiliam, 2010). This doctoral research focuses on the impacts of different modalities of formative assessment on students’ self-regulation in the specific context of IBSE. Two issues have guided this work. The first one, of empirical nature, aimed at analyzing the evaluation practices of teachers and their effects on the self-regulatory process. The second one was the development of a methodology to analyze the variables at stake. To do this, we proceeded by recording videos of class sessions and constructed an indicator grid which allowed us to analyze in situ process. Different teaching situations were compared. The first corresponded to formative assessments such as teachers implement it in their daily practices. The second concerned the assessment practices implemented by the same teachers the following year, after a workshop where teachers were invited to reflect on their practice. Our results show a better balance in the use of different formative assessment methods in the second situation, especially towards a greater empowerment of students and better taking account peers as resources. Students also demonstrate more efficient self-regulation of their behavior in the light of a greater time spent to produce solving strategies and a better commitment to the task. The question of how to formally integrate the self-assessment tool to student activity is still pending. Some ideas to effectively combine the evaluation with different learning tasks are thus proposed.

L’ère numérique change de plus en plus le monde en raison de la quantité de données récoltées chaque jour. Le domaine médical est fortement affecté par cette explosion, car l’exploitation de ces données est un véritable atout pour l’aide à la pratique médicale. Dans cette thèse, nous proposons d’utiliser les vidéos chirurgicales dans le but de créer un système de chirurgie assistée par ordinateur. Nous nous intéressons principalement à reconnaître les gestes chirurgicaux à chaque instant afin de fournir aux chirurgiens des recommandations et des informations pertinentes. Pour ce faire, l’objectif principal de cette thèse est de reconnaître les outils chirurgicaux dans les vidéos de chirurgie de la cataracte. Dans le flux vidéo du microscope, ces outils sont partiellement visibles et certains se ressemblent beaucoup. Pour relever ces défis, nous proposons d'ajouter une caméra supplémentaire filmant la table opératoire. Notre objectif est donc de détecter la présence des outils dans les deux types de flux vidéo : les vidéos du microscope et les vidéos de la table opératoire. Le premier enregistre l'oeil du patient et le second enregistre les activités de la table opératoire. Deux tâches sont proposées pour détecter les outils dans les vidéos de la table : la détection des changements et la détection de présence d'outil. Dans un premier temps, nous proposons un système similaire pour ces deux tâches. Il est basé sur l’extraction des caractéristiques visuelles avec des méthodes de classification classique. Il fournit des résultats satisfaisants pour la détection de changement, cependant, il fonctionne insuffisamment bien pour la tâche de détection de présence des outils sur la table. Dans un second temps, afin de résoudre le problème du choix des caractéristiques, nous utilisons des architectures d’apprentissage profond pour la détection d'outils chirurgicaux sur les deux types de vidéo. Pour surmonter les défis rencontrés dans les vidéos de la table, nous proposons de générer des vidéos artificielles imitant la scène de la table opératoire et d’utiliser un réseau de neurones à convolutions (CNN) à base de patch. Enfin, nous exploitons l'information temporelle en utilisant un réseau de neurones récurrent analysant les résultats de CNNs. Contrairement à notre hypothèse, les expérimentations montrent des résultats insuffisants pour la détection de présence des outils sur la table, mais de très bons résultats dans les vidéos du microscope. Nous obtenons des résultats encore meilleurs dans les vidéos du microscope après avoir fusionné l’information issue de la détection des changements sur la table et la présence des outils dans l’oeil. / The digital era is increasingly changing the world due to the sheer volume of data produced every day. The medical domain is highly affected by this revolution, because analysing this data can be a source of education/support for the clinicians. In this thesis, we propose to reuse the surgery videos recorded in the operating rooms for computer-assisted surgery system. We are chiefly interested in recognizing the surgical gesture being performed at each instant in order to provide relevant information. To achieve this goal, this thesis addresses the surgical tool recognition problem, with applications in cataract surgery. The main objective of this thesis is to address the surgical tool recognition problem in cataract surgery videos.In the surgical field, those tools are partially visible in videos and highly similar to one another. To address the visual challenges in the cataract surgical field, we propose to add an additional camera filming the surgical tray. Our goal is to detect the tool presence in the two complementary types of videos: tool-tissue interaction and surgical tray videos. The former records the patient's eye and the latter records the surgical tray activities.Two tasks are proposed to perform the task on the surgical tray videos: tools change detection and tool presence detection.First, we establish a similar pipeline for both tasks. It is based on standard classification methods on top of visual learning features. It yields satisfactory results for the tools change task, howev-lateer, it badly performs the surgical tool presence task on the tray. Second, we design deep learning architectures for the surgical tool detection on both video types in order to address the difficulties in manually designing the visual features.To alleviate the inherent challenges on the surgical tray videos, we propose to generate simulated surgical tray scenes along with a patch-based convolutional neural network (CNN).Ultimately, we study the temporal information using RNN processing the CNN results. Contrary to our primary hypothesis, the experimental results show deficient results for surgical tool presence on the tray but very good results on the tool-tissue interaction videos. We achieve even better results in the surgical field after fusing the tool change information coming from the tray and tool presence signals on the tool-tissue interaction videos.

Introdução:Tão importante quanto saber \"o que\" está mudando no desenvolvimento motor, é \"como\" e \"quando\" ocorrem essas mudanças. A falta desses dados ainda não permite a descrição desse processo de forma linear e contínua em grupos com distúrbios de movimento, tais como recém-nascidos com síndromede Down. A memória pode ser um fator de diferenciação no caso de desempenho motor e a construção de aquisição da trajetória do movimento. Portanto, aintegraçãomotora-cognitiva,e a ativação mnemônica poderiamser observadas na apresentação de uma riqueza do repertório motor demovimentação espontânea dos lactentes. Objetivo geral:Avaliara qualidade da movimentação espontânea em lactentescom síndrome de Down. Objetivos específicos: a) descrever as características genéticas e ambientais sobreas características deste grupo delactentes e o atual processo deintervenção; b) analisar a aplicabilidade da avaliaçãoGeneral Movementsem lactentes até oscinco meses de vidaem paísesde baixae médiarenda e; c) avaliar a movimentação espontânea como facilitadora do vínculo mãe-bebê. Métodos:revisão e elaboração da temática,por busca em bases de dados científicos, para estruturar de forma factível a execução da avaliação e aplicação do questionário neste grupo de lactentes. Foi um estudo exploratório,com a avaliaçãopresencial e por gravações de vídeo, de 47 lactentes, menores de 5 mesescom síndrome de Down pela avaliação dos General Movements; além do preenchimento de um questionário cominformaçõestais como: o momento em que o diagnóstico foidito para os pais, o tempo de amamentação, o período que permaneceram na unidade de Cuidados Intensivos, se os pais estavamempregados e a idade dos pais. A avaliação foi feita na idade Fidgety(de 11 a 20 semanas após a idade do termo). Os locais de coleta dos dados foram: 24 lactentes doHospital Infantil Darcy Vargas e a Instituição APAE de São Paulo (o hospital pertence ao SUS -Sistema Único de Saúde),ambos representam locais de referência no atendimento de crianças com síndrome 7de Downno Brasil, e 23 lactentes do banco de dados por vídeo do Centro de Estudos de General Movements da Áustria. Resultados:O escore da avaliação foi significativamente menor do que em lactentescom desfecho neurológico comum. Quatorze lactentes com Síndrome de Down apresentaram Fidgety Movementsnormais, 13não apresentaram FMe 20 apresentaram FM exagerados, muito rápidos ou muito lentos. A falta de movimentos na linha média e várias posturas atípicas foram observadas. Nem o parto prematuro nem a cardiopatia congênita estavam associados apresença dos FM oua mobilidade reduzida. Conclusões:Para a indicação do uso dessa avaliação de baixo custo no Brasil:a observação motora contribui para a avaliaçãoqualitativado movimento global efuncional dolactentede uma maneira assertiva, rápida e não invasiva.Quanto arelevância daavaliação da qualidade dosmovimentosglobaisemlactentes com síndrome de Down:aheterogeneidade nos FM esuas características peculiares indicamque a intervenção deve serfeita o quanto antes para estimular e aprimorar o repertoriomotor.Os dados apontam que a identificação dos FM em crianças com síndrome de Down, pode ser um marcador clínico para o planejamento e intervenção fisioterapêutica singular; contudo sua ausência não pode ser utilizada como indicador de normalidade motorafuncionalou pretexto para adiarintervenções clínicas. Uma segunda conclusão aponta para inexistência de associação entre cardiopatias congênitas e FM em crianças com síndrome de Down. / Introduction-Asimportant as knowing \"what\" is changing in motor development, it is \"how\" and \"when\" these changes occur. The lack of these data still does not allow the description of this process in a linear and continuous way in groups with movement disorders, such as Down syndrome newborns. The memory can be a factor of differentiation in the case of motor performance and the construction of acquisition of the trajectory of the movement. Therefore, integration and its complete activation can be observed in the presentation of a rich motor repertoire in the spontaneous movement of infants. General purpose-To assessthe quality of spontaneous movement in infants with Down syndrome. Specific purposes-a) to describe information about the characteristics of this group of infants and the process of intervention in low-income countries; B) to analyze the applicability of the General Movements instrument in infants up to five months of life in low-and middle-income countries; C) to assess the spontaneous movement as facilitator of the mother-baby bond. Methods-review and elaboration of the thematic by searching in scientific databases at a first moment. Subsequent there was an exploratory study was conducted with the evaluation of 47 infants with Down syndrome by General Movements; in addition to filling out a questionnaire of information such as: the time the diagnosis is told to the parents, the time of breastfeeding, the period they stayed in the Intensive Care unit, the parents are employed, and the age of the parents. The evaluation was done at the Fidgety age (from 11 to 20 weeks after the term age). The data collection sites were: 24 infants from Darcy Vargas Children\'s Hospital and the APAE Institution of São Paulo (the hospital belongs to SUS -Sistema Único de Saúde),both represent reference sites for the care of children with Down syndrome, and 23infants from Austria data recorded. Results-The assessment score was significantly lower than in infants with a common neurological outcome. Fourteen infants with Down syndrome had normal FM, 13 had no FM and 20 had FM exaggerated, very fast or very slow. The 10lack of movements for the midline and several atypical postures were observed. Neither preterm birth nor congenital heart disease were related to FM or reduced mobility. Conclusions-for the indication in the use of the evaluation in Brazil:the researchers demonstrated that the quality motor observation contributes to the functional evaluation of the young nervous system. The application of GM assessment to vulnerable populations such as Brazil is therefore highly recommended. For the relevance of the evaluation applied to this group of infants with Down\'s syndrome:the research showsthat the heterogeneity in FM and its peculiar characteristicsjustify the early intervention.

In the form of teacher didactical design research, this work addresses a didactical issue encountered during physics teaching in a Swedish upper secondary school. A need for renewed practical laboratory work related to Newtonian mechanics is met by proposing and designing an activity based on high- speed photography using the nowadays omnipresent smartphone, thus bringing new technology into the classroom. The activity – video analysis of the collision physics of football kicks – is designed and evaluated by following a didactical design cycle. The work elaborates on how the proposed laboratory activity relates to the potential and complications of experimental activities in science education, as described in the vast literature on the topic. It is argued that the use of smartphones constitutes an interesting use of new technology for addressing known problems of practical work. Of particular interest is that smartphones offer a way to bridge the gap between the everyday life of students and the world of physics experiments (smartphones are powerful pocket laboratories). The use of smartphones also avoids using unfamiliar laboratory equipment that is known to hinder focus on intended content, while at the same time exploring a powerful tool for data acquisition and analysis. Overall, the use of smartphones (and computers) in this manner can be seen as the result of applying Occam’s razor to didactics: only familiar and readily available instrumentation is used, and skills learned (movie handling and image analysis) are all educationally worthwhile. Although the activity was judged successful, a systematic investigation of learning outcome was out of scope. This means that no strong conclusions can be drawn based on this limited work. Nonetheless, the smartphone activity was well received by the students and should constitute a useful addition to the set of instructional approaches, especially since variation is known to benefit learning. The main failure of the design was an overestimation of student prior knowledge on motion physics (and its application to image data). As a consequence, the activity took required more time and effort than originally anticipated. No severe pitfalls of smartphone usage were identified, but it should be noted that the proposed activity – with its lack of well-defined results due to variations in kick strength – requires that the teacher is capable of efficiently analysing multiple student films (avoiding the feedback process to become overwhelmingly time consuming). If not all student films are evaluated, the feedback to the students may become of low quality, and misconceptions may pass under the radar. On the other hand, given that programming from 2018 will become compulsory, an interesting development of the activity would be to include handling of images and videos using a high-level programming language like Python.

Plankton microorganisms constitute the base of the marine food web and play a great role in global atmospheric carbon dioxide drawdown. Moreover, being very sensitive to any environmental changes they allow noticing (and potentially counteracting) them faster than with any other means. As such they not only influence the fishery industry but are also frequently used to analyze changes in exploited coastal areas and the influence of these interferences on local environment and climate. As a consequence, there is a strong need for highly efficient systems allowing long time and large volume observation of plankton communities. This would provide us with better understanding of plankton role on global climate as well as help maintain the fragile environmental equilibrium. The adopted sensors typically provide huge amounts of data that must be processed efficiently without the need for intensive manual work of specialists. A new system for general purpose particle analysis in large volumes is presented. It has been designed and optimized for the continuous plankton monitoring problem; however, it can be easily applied as a versatile moving fluids analysis tool or in any other application in which targets to be detected and identified move in a unidirectional flux. The proposed system is composed of three stages: data acquisition, targets detection and their identification. Dedicated optical hardware is used to record images of small particles immersed in the water flux. Targets detection is performed using a Visual Rhythm-based method which greatly accelerates the processing time and allows higher volume throughput. The proposed method detects, counts and measures organisms present in water flux passing in front of the camera. Moreover, the developed software allows saving cropped plankton images which not only greatly reduces required storage space but also constitutes the input for their automatic identification. In order to assure maximal performance (up to 720 MB/s) the algorithm was implemented using CUDA for GPGPU. The method was tested on a large dataset and compared with alternative frame-by-frame approach. The obtained plankton images were used to build a classifier that is applied to automatically identify organisms in plankton analysis experiments. For this purpose a dedicated feature extracting software was developed. Various subsets of the 55 shape characteristics were tested with different off-the-shelf learning models. The best accuracy of approximately 92% was obtained with Support Vector Machines. This result is comparable to the average expert manual identification performance. This work was developed under joint supervision with Professor Rubens Lopes (IO-USP). / Microorganismos planctônicos constituem a base da cadeia alimentar marinha e desempenham um grande papel na redução do dióxido de carbono na atmosfera. Além disso, são muito sensíveis a alterações ambientais e permitem perceber (e potencialmente neutralizar) as mesmas mais rapidamente do que em qualquer outro meio. Como tal, não só influenciam a indústria da pesca, mas também são frequentemente utilizados para analisar as mudanças nas zonas costeiras exploradas e a influência destas interferências no ambiente e clima locais. Como consequência, existe uma forte necessidade de desenvolver sistemas altamente eficientes, que permitam observar comunidades planctônicas em grandes escalas de tempo e volume. Isso nos fornece uma melhor compreensão do papel do plâncton no clima global, bem como ajuda a manter o equilíbrio do frágil meio ambiente. Os sensores utilizados normalmente fornecem grandes quantidades de dados que devem ser processados de forma eficiente sem a necessidade do trabalho manual intensivo de especialistas. Um novo sistema de monitoramento de plâncton em grandes volumes é apresentado. Foi desenvolvido e otimizado para o monitoramento contínuo de plâncton; no entanto, pode ser aplicado como uma ferramenta versátil para a análise de fluídos em movimento ou em qualquer aplicação que visa detectar e identificar movimento em fluxo unidirecional. O sistema proposto é composto de três estágios: aquisição de dados, detecção de alvos e suas identificações. O equipamento óptico é utilizado para gravar imagens de pequenas particulas imersas no fluxo de água. A detecção de alvos é realizada pelo método baseado no Ritmo Visual, que acelera significativamente o tempo de processamento e permite um maior fluxo de volume. O método proposto detecta, conta e mede organismos presentes na passagem do fluxo de água em frente ao sensor da câmera. Além disso, o software desenvolvido permite salvar imagens segmentadas de plâncton, que não só reduz consideravelmente o espaço de armazenamento necessário, mas também constitui a entrada para a sua identificação automática. Para garantir o desempenho máximo de até 720 MB/s, o algoritmo foi implementado utilizando CUDA para GPGPU. O método foi testado em um grande conjunto de dados e comparado com a abordagem alternativa de quadro-a-quadro. As imagens obtidas foram utilizadas para construir um classificador que é aplicado na identificação automática de organismos em experimentos de análise de plâncton. Por este motivo desenvolveu-se um software para extração de características. Diversos subconjuntos das 55 características foram testados através de modelos de aprendizagem disponíveis. A melhor exatidão de aproximadamente 92% foi obtida através da máquina de vetores de suporte. Este resultado é comparável à identificação manual média realizada por especialistas. Este trabalho foi desenvolvido sob a co-orientacao do Professor Rubens Lopes (IO-USP).

<div>The demand for global video has been burgeoning across industries. With the expansion and improvement of video-streaming services, cloud-based video is evolving into a necessary feature of any successful business for reaching internal and external audiences. Over-the-top (OTT) video streaming, e.g., Netfix and YouTube, has been dominating the global IP traffic in recent years. More than 50% of OTT video traffic are now delivered through content distribution networks (CDNs). Even though multiple solutions have been proposed for improving congestion in the CDN system, managing the ever-increasing traffic requires a fundamental understanding of the system and the different design fexibilities (control knobs) to make the best use of the hardware limitations. In Addition, there is no analytical understanding for the key quality of experience (QoE) attributes (stall duration, average quality, etc.) for video streaming when transmitted using CDN-based multi-tier infrastructure, which is the focus of this thesis. The key contribution of this thesis is to provide a white-box analytical understanding of the key QoE attributes of the enduser in cloud storage systems, which can be used to systematically address the choppy user experience and have optimized system designs. The rst key design involves the scheduling strategy, that chooses the subset of CDN servers to obtain the content. The second key design involves the quality of each video chunk. The third key design involves deciding which contents to cache at the edge routers and which content needs to be stored at the CDN. Towards solving these challenges, this dissertation is divided into three parts. Part 1 considers video streaming over distributed systems where the video segments are encoded using an erasure code for better reliability. Part 2 looks at the problem of optimizing the tradeoff between quality and stall of the streamed videos. In Part 3, we consider caching partial contents of the videos at the CDN as well as at the edge-routers to further optimize video streaming services. We present a model for describing a today's representative multi-tier system architecture</div><div>for video streaming applications, typically composed of a centralized origin server, several CDN sites and edge-caches. Our model comprehensively considers the following factors: limited caching spaces at the CDN sites and edge-routers, allocation of CDN for a video request, choice of different ports from the CDN, and the central storage and bandwidth allocation. With this model, we optimize different quality of experience (QoE) measures and present novel, yet efficient, algorithms to solve the formulated optimization problems. Our extensive simulation results demonstrate that the proposed algorithms signicantly outperform the state-of-the-art strategies. We take one step further and implement a small-scale video streaming system in a real cloud environment, managed by Openstack, and validate our results</div>

This doctoral thesis is concerned with multi-participant interaction systems of school lessons in a seventh grade comprehensive school German class. The multi-participant factor is assumed to be crucial for the school lesson and thus for its analysis not only because the mere presence and specific ratification of school lesson participants constitute the very event in general. It is, moreover, argued that each participant is a co-producer of the social event as it emerges, develops, and turns out, and that only the cooperation of all the participants ensures the successful execution of the social event. With a conversation-analytical approach, the mechanisms of multi-participant cooperation in the classroom are primarily investigated with regard to all participants’ individual strategies of coping with the requirements of being a student or being a teacher during lesson discourses, which constitute the major part of the school lessons under scrutiny. Based on this, participation in classroom interaction can be described along the lines of leading and supporting activities conditioned by mutual attestations of inconspicuousness.