資訊的進步讓學習者可透過多媒體進行動作學習，有偵測錯誤能力才能在無指導者的狀況下進行自我調整。本研究目的為探討多媒體教學影片講解順序對於學習者錯誤偵測能力的影響。研究對象為四年制大專學生，且無網球或其他持拍類運動背景60名 (男：33位；女：27位；年齡：22.0±1.0)。研究中預先設計A、B兩段網球正拍教學影片，A.先講解身體重心後講解球拍軌跡，以及B.先講解球拍軌跡後講解身體重心，實驗進行前參與者隨機觀看教學影片，建立認知基模，進行實驗時使用SMI 公司的RED遠距測量眼動儀，捕捉參與者偵測初學者網球正拍動作時的凝視軌跡，並以手機錄音的方式紀錄參與者口述偵測的錯誤，所得資料以獨立樣本t檢定處理分析 (α=.05)。結果顯示：1. 觀看不同順序的參與者凝視初學者正拍動作的視覺分配不同，其中手臂、軀幹的AOI達顯著差異；2. 分期階段的拉拍期手臂與軀幹的AOI達顯著差異；3. 擊球期與完成期則皆只有手臂AOI達顯著差異；4. 偵測錯誤結果上觀看A教學影片的參與者偵測較多錯誤，結果也達顯著差異。本研究結果發現不同順序教學影片對參與者偵測錯誤的視覺軌跡或能力皆有影響，建議未來網球正拍教學過程先給予身體重心的重點，將有利於學習者偵測錯誤能力提昇。 / The advent of information technology made learning motor skill through multimedia possible and error-detection ability is essential to self-regulate learning under no instructor condition. In order to study the effect of multimedia instructional video content sequence to the learner’s error-detection ability, this research used sixty (male:33; female:27; age: 22.0±1.0 yrs old) four-years university students with no tennis or racket sports experience as subjects. Two demonstration videos were designed: A: explain body Center of Gravity (CG) first follow by racket trajectory; B: explain racket trajectory first follow by body CG. Participants were randomly assigned to watch the videos to build-up their cognitive schema. The SMI RED remote eye tracker is used to capture participant’s gaze trajectory when watching the beginner’s Tennis forehand drive motion. A cellphone is used to record the participant’s dictation of detection of errors. Data gathered were analysed with independent t-test (α=.05). The results indicate that: 1. participants watching different video content sequence allocate their visual distribution differently, with the arm and truck AOI significantly different; 2. The AOI of arm and truck in backswing stage were significantly different; 3. In waving and follow through stage only the arm AOI exhibit significant difference. 4. In error-detection, watching the A instructional video group exhibit significant error-detection results. The results indicate instructional video content sequence affect the learner’s error-detection visual trajectory and ability. The author suggests that emphasize on body CG in tennis forehand drive instruction process can be favourable to the error-detection ability. / 第壹章 緒論 1 第一節 研究背景 1 第二節 研究目的 2 第三節 研究範圍與限制 2 第四節 操作性定義 3 第貳章 文獻探討 6 第一節 多媒體與學習理論 7 第二節 動作學習理論 14 第三節 眼動與運動相關研究 22 第四節 網球正拍動作 25 第五節 本章總結 27 第參章 研究方法 28 第一節 研究架構 28 第二節 研究對象 29 第三節 實驗時間與地點 29 第四節 教學影片與實驗素材 30 第五節 實驗儀器 32 第六節 場地布置 33 第七節 實驗步驟 34 第八節 收集資料與處理方法 40 第九節 統計分析方法 42 第十節 預備實驗 43 第肆章 結果與討論 45 第一節 凝視AOI區域分佈 46 第二節 偵測錯誤語音資料 51 第三節 偵測錯誤凝視順序 54 第四節 綜合討論 55 第伍章 結論與建議 58 第一節 結論 58 第二節 建議 58 參考文獻 60 附件一 正拍動作技術檢核表 67 附件二 人體研究倫理委員會通過證明 68 附件三 受試者同意書 69

Denna studie undersöker hur minskad anonymitet kan förändra sättet människor skriver kommentarer på YouTube-videor. YouTube begärde nyligen att användarna skulle ange sina fullständiga namn. Användare är ännu inte skyldiga att ändra sitt användarnamn, men om de inte gör det ombes de att svara på varför de väljer att ha kvar sitt alias. För att undersöka detta lät vi först användare svara på en enkät för att se vad de tycker om att vara anonym kontra att ge ut sitt riktiga namn, och om att ge sitt riktiga namn gör dem mer uppmärksamma kring deras sätt att kommentera på YouTube. Sedan, för att se vad som verkligen händer i kommentarsfälten, gick vi igenom ett antal kommentarer för att hitta mönster i hur ett alias kontra ett riktigt namn relaterar till processen att skriva negativa kontra positiva kommentarer. Kommentarerna analyserades med hjälp av en modell som mäter olika grader av negativitet och positivitet i en kommentar. Våra studier visar att ändringen från ett alias till riktigt namn, påverkar hur kommentarer skrivs och att ett högre antal negativa kommentarer kommer från användare som använder ett alias. Resultatet är en indikation på hur minskning av anonymitet kan leda till mindre negativitet i YouTubes kommentarsfält. / This study examines how reducing anonymity options can change the way people write comments on YouTube-videos. YouTube recently requested users to enter their real names as a username. Users are not yet obligated to do so, but if they do not, they are asked to answer why they chose remain their alias. To investigate this we first let users answer a survey to see what they felt about being anonymous versus giving their real name, and if providing their real name made them more cautious about their way of interacting on YouTube. Then, to see what really is going on in the comment sections, we went through a group of comments to spot patterns in how an alias or a real name is related to the process of writing negative versus positive comments. Comments were analyzed with help from a model that measures the stages of negativity and positivity in a comment. Our studies indicate that the change from alias to real name, make an impact in how comments are made and that much higher rate of negative comments come from users using an alias. The result is an indication that less anonymity may lead to less negativity in YouTube’s comment sections.

Acompanha: Desmistificando a matemática por meio da estatística / A utilização das mídias digitais é uma realidade em todos os meios sociais, e não deveria ser diferente na Educação. No entanto, em função da necessidade de mudanças por parte dos docentes e das escolas, a implementação do uso de tecnologias em salas de aula ainda é pequena. Neste contexto, o propósito deste trabalho é apresentar uma análise sobre o ensino dos conteúdos de Estatística, pertencentes à disciplina de Matemática, ministrados no 7º ano do Ensino Fundamental da Educação Básica. O estudo foi realizado em uma escola pública do município de Cambé, buscando identificar a percepção dos alunos em relação aos conteúdos estudados e ao uso de recursos midiáticos nas aulas de estatística. A metodologia adotada foi de caráter descritivo quanti/qualitativo e contou com pesquisa bibliográfica e uma pesquisa de campo realizada no âmbito da escola objeto do estudo. Os recursos digitais inseridos nas aulas expositivas foram videoaulas preparadas especificamente com cada conteúdo e inseridas em um vlog de estatística. Por meio da análise dos questionários aplicados, das gravações e das anotações feitas durante a aplicação do trabalho, pôde-se constatar que os alunos que participaram da pesquisa tiveram uma boa participação nas aulas, ocorrendo melhorias no aprendizado e na compreensão dos conteúdos de alguns pontos de estatística. Concluindo, importante ressaltar que é necessário o professor estruturar os conteúdos das aulas em uma ordem cronológica. Primeiramente preparar os alunos com uma exposição teórica, em seguida, efetuar a inserção do vlog sobre o assunto, por último, finalizar a aula com um fechamento sanando as dúvidas dos alunos. Dessa forma, os alunos poderão perceber uma associação da teoria com a prática vivenciada por eles no seu dia a dia extraclasse. / The use of digital media is a reality in all social media, and should not be different in Education. However, due to the need for changes on the part of teachers and schools, implementation of the use of technology in classrooms is still small. In this context, the purpose of this article is to present an analysis on the teaching of the contents of Statistics belonging to Mathematics, taught in the 7th year of Basic Education. The study was carried out in a public school, in the city of Cambé, seeking to identify the students' perception regarding the contents studied and the use of resources media technology in the statistics classes. The methodology adopted was quantitative / qualitative descriptive and counted on research Bibliography and a field research carried out within the scope of the school object of the study. The digital resources inserted in the lectures were statistical vlogs prepared specifically for each class. The results obtained in the investigation showed that there was a high motivation of the students to attend and participate in the classes. There are improvements in learning in the understanding of the contents of statistics in the daily school life. In conclusion, it is important to emphasize that it is necessary for the teacher to structure the contents of the classes in a chronological order. First, prepare the students with a theoretical exposition, then effect the insertion of the vlog on the subject and finally, finish the class with a closing answering the doubts of the students. In this way, the students will be able to realize an association of the theory with the practice experienced by them in their day to day extraclass.

[pt] A detecção e o subsequente rastreamento de objetos em sequencias de vídeo é um desafio no que tange o processamento de vídeos em tempo real. Nesta tese propomos um método de detecção em tempo real adequado para o processamento de vídeos de alta definição. Neste método utilizamos um procedimento de segmentação de quadros usando as imagens integrais de frente, o que permite o rápido descarte de várias partes da imagem a cada quadro, desta maneira atingindo uma alta taxa de quadros processados por segundo. Estendemos ainda o algoritmo proposto para que seja possível detectar múltiplos objetos em paralelo. Além disto, através da utilização de uma GPU e técnicas que podem ter seu desempenho aumentado por meio de paralelismo, como o operador prefix sum, conseguimos atingir um desempenho ainda melhor do algoritmo, tanto para a detecção do objeto, como na etapa de treinamento de novas classes de objetos. / [en] The detection and subsequent tracking of objects in video sequences is a challenge in terms of video processing in real time. In this thesis we propose an detection method suitable for processing high-definition video in real-time. In this method we use a segmentation procedure through integral image of the foreground, which allows a very quick disposal of various parts of the image in each frame, thus achieving a high rate of processed frames per second. Further we extend the proposed method to be able to detect multiple objects in parallel. Furthermore, by using a GPU and techniques that can have its performance enhanced through parallelism, as the operator prefix sum, we can achieve an even better performance of the algorithm, both for the detection of the object, as in the training stage of new classes of objects.

A presente investigação pretende, de forma qualitativa, compreender se os vídeos indexados podem modificar o planejamento pedagógico do professor de escola municipal de Porto Alegre. Para atingir esse objetivo, buscamos embasamento em uma fundamentação teórica de fundo piagetiano, voltada para a construção do conhecimento e suas diversas possibilidades, além das teorias relativas às questões do vídeo e do computador no processo educativo, a partir do seu envolvimento como suporte pedagógico orientado na pedagogia de projetos. Inserido nesse contexto, propõe-se o SACCA (Sistema Automático de Catalogação de Conteúdo Audiovisual), que possibilita o processo de indexação de vídeos com a intermediação de um usuário que assiste ao vídeo e define as palavras-chaves que irão compor o índice. Posteriormente, poderá se navegar no interior desses vídeos permitindo ter uma visão rápida do que o vídeo trata. Foram acrescentados 49 vídeos da História do Brasil e disponibilizados para os professores de História do 3º Ciclo das Escolas da Rede Municipal de Porto Alegre participantes da pesquisa. Os dados foram coletados através de entrevistas. A partir das perguntas norteadoras da pesquisa e da investigação realizada, são definidas categorias de análise para as reflexões e as conclusões desta tese. Os resultados obtidos indicam que o uso dos vídeos indexados pode modificar o planejamento pedagógico do professor. / The present investigation intends to understand in a qualitative way if the indexed videos can modify the teacher's pedagogical planning of a public school in Porto Alegre. We searched for a theoretical Piagetian approach, in order to achieve this purpose, turned to the construction of knowledge and its several possibilities, besides the theories related to the points of the video and computer in the educational process, starting from its involvement as a pedagogical support guided in the pedagogy of projects. Inserted in this context, SACCA (Automatic System of Cataloguing of Audiovisual Content) is proposed, which makes the process of indexation of videos possible, with the user's intermediation who sees the video and defines the key-words, which will compose the index. Later, one can navigate in the interior of these videos allowing a fast vision of what the video deals with. 49 videos of the Brazilian History were added and available for the History teachers of the 3rd Cycle of the Public Schools in Porto Alegre, participants in the research. The data were collected through interviews. Starting from the directed questions of the research and of the accomplished investigation, analysis categories for the reflections and the conclusions of this theory are defined. The obtained results indicate that the use of the indexed videos can modify the teacher's pedagogical planning.

The production from an oil reservoir is a mixture of liquids (oil and water) and gas, and is often maintained by using a pump placed in the well to ensure a continuous flow to the surface. Electrical Submersible Pumps consist of stacked centrifugal pump stages, each comprising a bladed impeller (rotating part) and diffuser (stationary part). In multiphase conditions, the gas tends to accumulate in the impeller, severely reducing the pressure produced by the pump. Radial-flow pumps operate in a plane perpendicular to their rotation axis, while mixed-flow pumps are characterised by a lower meridional angle (generally 40 to 80 degrees), and are generally better at handling gas-liquid mixtures. We first describe the impact of gas on the whole pumping system, from the reservoir to the storage facility, and give context to the subject. The available literature shows that the size of the gas bubbles present in the fluid is critical to the pump performance. A transparent, full-scale pump was built in order to explore the flow features in single and multiphase flows. Laser Doppler Velocimetry and high speed imaging in single phase flow showed a high turbulence level in the wake of the impeller blades, and recirculation cells at low flow rates. In gas-liquid conditions, we demonstrated that the bubble size varies within a pump stage, as break-up occurs at the impeller tip, and coalescence is dominant in the diffuser, especially because of recirculation. The first impeller acted as a mixer, and at moderate to high gas fractions (10 to 30%), the flow patterns at the stage level alternated between bubbly and radially separated flows. Finally, a dispersed-gas model was developed to predict the pressure rise in a mixed-flow pump impeller under gas-liquid conditions. This model based on the forces acting on a single spherical gas bubble, was implemented with a simplified, parametric representation of the flow field in a mixed-flow impeller. In the meridional direction, the Coriolis force opposes the centrifugal force and the adverse pressure gradient. Both forces tend to retain the gas bubble within the impeller. The relative magnitude of the drag force strongly depends on the maximal bubble diameter, which was determined as a function of the flow conditions and used to calculate the gas velocity through the impeller. This method resulted in a better agreement with the experimental data than a one-dimensional two-fluid model where the gas phase follows the same path as the liquid. We used the dispersed-gas model to give quantitative evidence that low blade and meridional angles reduce the gas accumulation and the associated performance degradation.

Submitted by MILENA APARECIDA VENDRAMINI SATO null (mivendramini1@hotmail.com) on 2016-01-20T00:32:25Z No. of bitstreams: 1 Dissertação.pdf: 3350747 bytes, checksum: 74d02ac066b77cb3ed5200686edb266f (MD5) / Approved for entry into archive by Sandra Manzano de Almeida (smanzano@marilia.unesp.br) on 2016-01-20T17:25:21Z (GMT) No. of bitstreams: 1 sato_mav_me_bauru.pdf: 3350747 bytes, checksum: 74d02ac066b77cb3ed5200686edb266f (MD5) / Made available in DSpace on 2016-01-20T17:25:21Z (GMT). No. of bitstreams: 1 sato_mav_me_bauru.pdf: 3350747 bytes, checksum: 74d02ac066b77cb3ed5200686edb266f (MD5) Previous issue date: 2015-12-17 / Não recebi financiamento / A pesquisa, vinculada à linha Tecnologias da Informação e da Comunicação na Educação Básica, do Programa de Pós-Graduação-Mestrado em Docência para Educação Básica da Faculdade de Ciências da Universidade Estadual Paulista- UNESP, surgiu diante de indagações enquanto docente dos anos iniciais do Ensino Fundamental, no que tange a questão do uso das Tecnologias Digitais da Informação e Comunicação (TDIC) pelos alunos como entretenimento e ou possibilidades de aprendizagem. Estudos realizados apontaram que, atualmente, faz-se uso das tecnologias como diversão sem a devida exploração das suas possibilidades de aprendizagem, fato este que deve ser pauta de discussão nas escolas pelos professores. Assim, a pesquisa teve como objetivo explorar as possibilidades pedagógicas do uso das TDIC nos anos iniciais do Ensino Fundamental. Optamos por uma pesquisa de natureza qualitativa, do tipo experimental e participante para o desenvolvimento de atividades práticas, como a construção de vídeos digitais. Para isso, realizamos uma sequência didática com 23 alunos, do 2˚ ano do Ensino Fundamental, de uma escola do Sistema Municipal de Ensino, localizada no interior do estado de São Paulo, em que a professora da turma é também a pesquisadora. Efetuamos, portanto, o seguinte percurso metodológico: 1) revisão bibliográfica; 2) entrevista semiestruturada com os alunos, antes da criação do audiovisual, para verificarmos o nível de compreensão em relação aos vídeos digitais; 3) construção e realização de sequência didática; 4) entrevista semiestruturada com os alunos para avaliar as atividades desenvolvidas; 5) análise e interpretação dos dados; 6) elaboração do produto final. Os dados alcançados nas entrevistas foram apresentados num quadro e, posteriormente, analisados à luz das teorias exploradas na pesquisa, bem como, as atividades pedagógicas realizadas pelos alunos. A análise dos dados da entrevista inicial apontou que os alunos assistem vídeos com o uso de diferentes recursos tecnológicos, utilizam as tecnologias (celular, tablet) para o divertimento, filmam momentos do cotidiano e compartilham na Internet. Em continuidade, analisamos o material obtido pela sequência didática que mostrou a apropriação de conhecimentos pelos alunos para a produção de vídeos digitais por meio da ajuda de sujeitos mais experientes e, por fim, verificamos na entrevista final que os participantes se apropriaram dos conteúdos curriculares, bem como, da função social das tecnologias. Acreditamos que a criação de vídeos digitais possa contribuir no processo de aprendizagem do aluno, podendo ser vista como uma ferramenta tecnológica capaz de informar e comunicar situações e problemas que fazem parte da realidade do aluno-autor. E, por fim, proporcionar o uso consciente das TDIC. Devido às especificidades do Mestrado Profissional apresentamos um produto final que consistiu na elaboração da sequência didática. / The research, linked to Information Technologies e Communication in Basic Education re-search line, from the Pos-Graduation Program, Masters Teaching Degree for Basic Education in the Science College of the Universidade Estadual Paulista – UNESP, has come to existence starting from questions about teachers from the early years of Elementary School, about the usage of Digital Information and Communication Technologies (DICT) by the students as entertainment and/or learning possibilities. Recent studies point to these technologies being used for fun, without exploring its learning possibilities, which should be discussed in schools by the teachers. Therefore, the research had as objectives to explore the pedagogical possibili-ties of using DICT in the early years of Elementary School. We choose a qualitative research, of the experimental and participative kind to develop practical activities, such as the creation of digital videos. To do so, we performed a didactical sequence with 23 students, of the 2º year of Elementary School, from a Municipal School, on upstate of São Paulo. The class teacher is also the researcher. We also followed these methodological procedures: 1) Biblio-graphic review; 2)Semi structured interview with the students, before creating the video, checking the comprehension level about digital videos; 3) Construction and application of the didactical sequence; 4) Data interpretation and analysis; 6) Final product elaboration. The collected data were presented on a chart and, afterwards, analyzed under the light of the theo-ries utilized on this research, as well as the pedagogical activities performed by the students. The data analysis of the initial interview pointed that the students utilize different kind of technologies to watch videos, that they utilize these technologies (cellphones, tablets) for en-tertainment, creating everyday videos and sharing them on the Internet. We then analyzed the material from the didactical sequence, which showed the knowledge appropriation of the stu-dents for the digital video production through the help of more experienced persons, and, at last, we checked in the final interview that the participants appropriated themselves of the curricular content and of the social function of technologies as well. We believe that the digi-tal video creation can contribute in the learning process, and can be seen as technological tool capable of inform e communicate situations and problems from the author-student reality. And, at last, to provide the conscious use of the DICT. Due to the specificities of the Profes-sional Masters Degree we present a final product which consisted in the elaboration of the didactical sequence.

Made available in DSpace on 2016-06-02T20:25:22Z (GMT). No. of bitstreams: 1 6221.pdf: 6384149 bytes, checksum: 3cd629d564c6cf7e30845046a65a3868 (MD5) Previous issue date: 2014-03-20 / Financiadora de Estudos e Projetos / Our research aims to undertake the survey of some representations of contemporary reading practices by analyzing videos to encourage reading posted on the YouTube website. For this analysis we conducted a survey of videos that, by their similarities and differences, allowed us to organize them into two categories (Campaigns and Protocampaigns), whose subdivision was based on our observation of aspects such as the motivation of its production, the strategies of their generic composition (gender campaign), the degree of specialization of its production, and the means of media movement for which these campaigns were originally made. These categories were also divided into two thematic areas, in our view, illustrating how reading is preponderantly represented in these media productions. They are: Who reads, progresses and Who reads, travels . By relying on principles of French Discourse Analysis of and Cultural History of Reading, we will develop our analysis in order to describe any discursive representations that make reading these videos aimed to promote it, as well as describe the representations of reader the producers of these videos make the audience they are targeting. Therefore, in our research we seek to analyze some evidence entered in the production of these videos, projections of reading practices, to reflect on a given imaginary we share today about what reading is, how it should be exercised, than being a good reader, and how to promote it. / Nossa pesquisa objetiva empreender o levantamento de algumas representações de práticas de leitura contemporâneas por meio da análise de vídeos de incentivo à leitura postados no site do YouTube. Para tal análise fizemos um levantamento dos vídeos que, por suas semelhanças e diferenças, permitiu-nos organizá-los em duas categorias (a das Campanhas e a das Protocampanhas), cuja subdivisão pautou-se em nossa observação de aspectos tais como a motivação de sua produção; as estratégias de sua composição genérica (gênero campanha); o grau de especialização de sua elaboração; e os meios midiáticos de circulação para os quais essas campanhas originalmente foram feitas. Tais categorias também foram divididas em dois eixos temáticos que, a nosso ver, ilustram o modo como a leitura é preponderantemente representada nessas produções audiovisuais. São eles: Quem lê, progride e Quem lê, viaja . Apoiando-nos em princípios da Análise do Discurso de linha francesa e da História Cultural da Leitura, desenvolveremos nossa análise de modo a descrever eventuais representações discursivas que se fazem da leitura nestes vídeos que visam promovê-la, assim como descrever as representações de leitor que os produtores desses vídeos fazem do público a que se dirigem. Portanto, buscamos em nossa pesquisa analisar alguns indícios, inscritos na produção desses vídeos, de projeções de práticas de leitura, de modo a refletir sobre um dado imaginário que partilhamos hoje em dia acerca do que é a leitura, de como ela deve ser exercida, do que é ser um bom leitor, e de como proceder para promovê-la.

Continuous improvements to the state of the art have made it easier to create, send and receive vast quantities of video over the Internet. Catalysed by these developments, video is now the largest, and fastest growing type of traffic on modern IP networks. In 2015, video was responsible for 70% of all traffic on the Internet, with an compound annual growth rate of 27%. On the other hand, concerns about the growing energy consumption of ICT in general, continue to rise. It is not surprising that there is a significant energy cost associated with these extensive video usage patterns. In this thesis, I examine the energy consumption of typical video configurations during decoding (playback) and encoding through empirical measurements on an experimental test-bed. I then make extrapolations to a global scale to show the opportunity for significant energy savings, achievable by simple modifications to these video configurations. Based on insights gained from these measurements, I propose a novel, energy-aware Quality of Experience (QoE) metric for digital video - the Energy - Video Quality Index (EnVI). Then, I present and evaluate vEQ-benchmark, a benchmarking and measurement tool for the purpose of generating EnVI scores. The tool enables fine-grained resource-usage analyses on video playback systems, and facilitates the creation of statistical models of power usage for these systems. I propose GreenDASH, an energy-aware extension of the existing Dynamic Adaptive Streaming over HTTP standard (DASH). GreenDASH incorporates relevant energy-usage and video quality information into the existing standard. It could enable dynamic, energy-aware adaptation for video in response to energy-usage and user ‘green' preferences. I also evaluate the subjective perception of such energy-aware, adaptive video streaming by means of a user study featuring 36 participants. I examine how video may be adapted to save energy without a significant impact on the Quality of Experience of these users. In summary, this thesis highlights the significant opportunities for energy savings if Internet users gain an awareness about their energy usage, and presents a technical discussion how this can be achieved by straightforward extensions to the current state of the art.

Cette thèse est motivée par le diagnostic, l’évaluation, la maintenance et la promotion de l’indépendance des personnes souffrant de maladies démentielles pour leurs activités de la vie quotidienne. Dans ce contexte nous nous intéressons à la reconnaissance automatique des activités de la vie quotidienne.L’analyse des vidéos de type égocentriques (où la caméra est posée sur une personne) a récemment gagné beaucoup d’intérêt en faveur de cette tâche. En effet de récentes études démontrent l’importance cruciale de la reconnaissance des objets actifs (manipulés ou observés par le patient) pour la reconnaissance d’activités et les vidéos égocentriques présentent l’avantage d’avoir une forte différenciation entre les objets actifs et passifs (associés à l’arrière plan). Une des approches récentes envers la reconnaissance des éléments actifs dans une scène est l’incorporation de la saillance visuelle dans les algorithmes de reconnaissance d’objets. Modéliser le processus sélectif du système visuel humain représente un moyen efficace de focaliser l’analyse d’une scène vers les endroits considérés d’intérêts ou saillants,qui, dans les vidéos égocentriques, correspondent fortement aux emplacements des objets d’intérêt. L’objectif de cette thèse est de permettre au systèmes de reconnaissance d’objets de fournir une détection plus précise des objets d’intérêts grâce à la saillance visuelle afin d’améliorer les performances de reconnaissances d’activités de la vie de tous les jours. Cette thèse est menée dans le cadre du projet Européen Dem@care.Concernant le vaste domaine de la modélisation de la saillance visuelle, nous étudions et proposons une contribution à la fois dans le domaine "Bottom-up" (regard attiré par des stimuli) que dans le domaine "Top-down" (regard attiré par la sémantique) qui ont pour but d’améliorer la reconnaissance d’objets actifs dans les vidéos égocentriques. Notre première contribution pour les modèles Bottom-up prend racine du fait que les observateurs d’une vidéo sont normalement attirés par le centre de celle-ci. Ce phénomène biologique s’appelle le biais central. Dans les vidéos égocentriques cependant, cette hypothèse n’est plus valable.Nous proposons et étudions des modèles de saillance basés sur ce phénomène de biais non central.Les modèles proposés sont entrainés à partir de fixations d’oeil enregistrées et incorporées dans des modèles spatio-temporels. Lorsque comparés à l’état-de-l’art des modèles Bottom-up, ceux que nous présentons montrent des résultats prometteurs qui illustrent la nécessité d’un modèle géométrique biaisé non-centré dans ce type de vidéos. Pour notre contribution dans le domaine Top-down, nous présentons un modèle probabiliste d’attention visuelle pour la reconnaissance d’objets manipulés dans les vidéos égocentriques. Bien que les bras soient souvent source d’occlusion des objets et considérés comme un fardeau, ils deviennent un atout dans notre approche. En effet nous extrayons à la fois des caractéristiques globales et locales permettant d’estimer leur disposition géométrique. Nous intégrons cette information dans un modèle probabiliste, avec équations de mise a jour pour optimiser la vraisemblance du modèle en fonction de ses paramètres et enfin générons les cartes d’attention visuelle pour la reconnaissance d’objets manipulés. [...] / The rationale and motivation of this PhD thesis is in the diagnosis, assessment,maintenance and promotion of self-independence of people with dementia in their InstrumentalActivities of Daily Living (IADLs). In this context a strong focus is held towardsthe task of automatically recognizing IADLs. Egocentric video analysis (cameras worn by aperson) has recently gained much interest regarding this goal. Indeed recent studies havedemonstrated how crucial is the recognition of active objects (manipulated or observedby the person wearing the camera) for the activity recognition task and egocentric videospresent the advantage of holding a strong differentiation between active and passive objects(associated to background). One recent approach towards finding active elements in a sceneis the incorporation of visual saliency in the object recognition paradigms. Modeling theselective process of human perception of visual scenes represents an efficient way to drivethe scene analysis towards particular areas considered of interest or salient, which, in egocentricvideos, strongly corresponds to the locus of objects of interest. The objective of thisthesis is to design an object recognition system that relies on visual saliency-maps to providemore precise object representations, that are robust against background clutter and, therefore,improve the recognition of active object for the IADLs recognition task. This PhD thesisis conducted in the framework of the Dem@care European project.Regarding the vast field of visual saliency modeling, we investigate and propose a contributionin both Bottom-up (gaze driven by stimuli) and Top-down (gaze driven by semantics)areas that aim at enhancing the particular task of active object recognition in egocentricvideo content. Our first contribution on Bottom-up models originates from the fact thatobservers are attracted by a central stimulus (the center of an image). This biological phenomenonis known as central bias. In egocentric videos however this hypothesis does not alwayshold. We study saliency models with non-central bias geometrical cues. The proposedvisual saliency models are trained based on eye fixations of observers and incorporated intospatio-temporal saliency models. When compared to state of the art visual saliency models,the ones we present show promising results as they highlight the necessity of a non-centeredgeometric saliency cue. For our top-down model contribution we present a probabilisticvisual attention model for manipulated object recognition in egocentric video content. Althougharms often occlude objects and are usually considered as a burden for many visionsystems, they become an asset in our approach, as we extract both global and local featuresdescribing their geometric layout and pose, as well as the objects being manipulated. We integratethis information in a probabilistic generative model, provide update equations thatautomatically compute the model parameters optimizing the likelihood of the data, and designa method to generate maps of visual attention that are later used in an object-recognitionframework. This task-driven assessment reveals that the proposed method outperforms thestate-of-the-art in object recognition for egocentric video content. [...]