O objetivo da tese é analisar a relação entre política e história na obra ciceroniana. Primeiramente, examinamos a concepção de homem, principalmente do sábio-político, e sua capacidade de ação na república. Analisamos a ação humana retratada nos diálogos filosóficos, nos discursos e nas narrativas históricas; observamos que a ação política é a matéria das narrativas históricas e os exemplos históricos são constitutivos da argumentação político-filosófica. Dessa maneira, as obras políticas são particularizadas pelos exemplos históricos, e as narrativas históricas são universalizadas pela presença do pensamento político-filosófico. Segundo a preceituação da narrativa histórica, o recurso à história tem, em todas as obras, uma função pedagógico-política ao guardar as ações dignas de memória do passado, falar ao tempo presente e poder ser estendido ao futuro. Por meio da análise do recurso à história, observamos uma concepção ciceroniana do curso dos acontecimentos em Roma, que não é nem circular nem linear. Mediante a análise das obras, percebemos que Cícero retrata a República de seu tempo, ou seja, do presente, como decadente, e apenas haveria expectativa de melhora se as ações exemplares do passado fossem recuperadas. / The aim of the thesis is to analyze the relation between politics and history in Ciceronian work. Initially, we examine the conception of man, especially the wise-politician, and his capacity for action in the republic. We analyze human action portrayed in philosophical dialogues, discourses and historical narratives; we observe that political action is the subject of historical narratives and historical examples are constitutive of political-philosophical argumentation. Thus, political works are particularized by historical examples and historical narratives are universalized by the presence of political-philosophical thought. According to the precept of historical narrative, resorting to history has in all works a pedagogical-political function, preserving the actions worthy of memory of the past, speaking to the present time and being able to be extended to the future. Through the analysis of the use of history we observe a Ciceronian conception of the course of history in Rome, which is neither circular nor linear. Through the analysis of the works we understand that Cicero describes the Republic of his time, that is, of the present, as decadent, and there would be only expectation of improvement if the exemplary actions of the past were recovered.

Traditional supervised visual recognition methods require a great number of annotated examples for each concerned class. The collection and annotation of visual data (e.g., images and videos) could be laborious, tedious and time-consuming when the number of classes involved is very large. In addition, there are such situations where the test instances are from novel classes for which training examples are unavailable in the training stage. These issues can be addressed by zero-shot learning (ZSL), an emerging machine learning technique enabling the recognition of novel classes. The key issue in zero-shot visual recognition is the semantic gap between visual and semantic representations. We address this issue in this thesis from three different perspectives: visual representations, semantic representations and the learning models. We first propose a novel bidirectional latent embedding framework for zero-shot visual recognition. By learning a latent space from visual representations and labelling information of the training examples, instances of different classes can be mapped into the latent space with the preserving of both visual and semantic relatedness, hence the semantic gap can be bridged. We conduct experiments on both object and human action recognition benchmarks to validate the effectiveness of the proposed ZSL framework. Then we extend the ZSL to the multi-label scenarios for multi-label zero-shot human action recognition based on weakly annotated video data. We employ a long short term memory (LSTM) neural network to explore the multiple actions underlying the video data. A joint latent space is learned by two component models (i.e. the visual model and the semantic model) to bridge the semantic gap. The two component embedding models are trained alternately to optimize the ranking based objectives. Extensive experiments are carried out on two multi-label human action datasets to evaluate the proposed framework. Finally, we propose alternative semantic representations for human actions towards narrowing the semantic gap from the perspective of semantic representation. A simple yet effective solution based on the exploration of web data has been investigated to enhance the semantic representations for human actions. The novel semantic representations are proved to benefit the zero-shot human action recognition significantly compared to the traditional attributes and word vectors. In summary, we propose novel frameworks for zero-shot visual recognition towards narrowing and bridging the semantic gap, and achieve state-of-the-art performance in different settings on multiple benchmarks.

A critical requirement for achieving ubiquity of artificial intelligence is to provide intelligent environments with the ability to recognize and understand human behaviour. If this is achieved, proactive interaction can occur and, more interestingly, a great variety of services can be developed. In this thesis we aim to support the development of ambient-assisted living services with advances in human behaviour analysis. Specifically, visual data analysis is considered in order to detect and understand human activity at home. As part of an intelligent monitoring system, single- and multi-view recognition of human actions is performed, along several optimizations and extensions. The present work may pave the way for more advanced human behaviour analysis techniques, such as the recognition of activities of daily living, personal routines and abnormal behaviour detection.

xv, 140 p. : ill. A print copy of this thesis is available through the UO Libraries. Search the library catalog for the location and call number. / Although research on action processing indicates people segment action according to a partonomic goal hierarchy, no previous research has investigated whether memory for complex human action is actually organized in the mind with respect to goals. This dissertation explored the primary organization of action memory in adults and young children and explored the role of familiarity in young children's organization of action in memory. Borrowing from the text memory literature, a priming experiment was designed to investigate the degree to which action memory is organized with respect to goals versus veridical temporal structure. In all studies, participants viewed videos in which goals were carried out in an interleaved fashion, such that the execution of a goal was at times interrupted by action related to the other goal. In a first experiment with adults, the results indicated that adults reorganize action information in memory in order to emphasize goal structure relative to verbatim temporal structure. A second control experiment with adults clarified that the goal priming effect observed in the first experiment arose as a result of viewing the action scenario and was not simply due to the stimuli cuing pre-existing semantic memory. The results of a third experiment with adults revealed this process of goal organization is unlikely to be a by-product of goal-based linguistic encoding, but instead reflects encoding of human action itself. Young children's action memory was examined in a fourth experiment, and the role of children's familiarity with the action scenarios in action memory organization was also explored. Children did not display a significant tendency to organize action according to goal inferences (or temporal structure, either, for that matter). As well, children's prior familiarity with action did not modulate their memory organization to any significant degree. Overall, the results of this dissertation suggest that adult memory for action emphasizes goal inferences but cannot speak to how or when this process in achieved developmentally. These findings have implications for contemporary theories of action processing, point to commonalities in the processing of language and human action, and open the door to future research into the development of action memory organization. / Committee in charge: Dare Baldwin, Chairperson, Psychology; Marjorie Taylor, Member, Psychology; Ulrich Mayr, Member, Psychology; Eric Pederson, Outside Member, Linguistics

O objetivo da tese é analisar a relação entre política e história na obra ciceroniana. Primeiramente, examinamos a concepção de homem, principalmente do sábio-político, e sua capacidade de ação na república. Analisamos a ação humana retratada nos diálogos filosóficos, nos discursos e nas narrativas históricas; observamos que a ação política é a matéria das narrativas históricas e os exemplos históricos são constitutivos da argumentação político-filosófica. Dessa maneira, as obras políticas são particularizadas pelos exemplos históricos, e as narrativas históricas são universalizadas pela presença do pensamento político-filosófico. Segundo a preceituação da narrativa histórica, o recurso à história tem, em todas as obras, uma função pedagógico-política ao guardar as ações dignas de memória do passado, falar ao tempo presente e poder ser estendido ao futuro. Por meio da análise do recurso à história, observamos uma concepção ciceroniana do curso dos acontecimentos em Roma, que não é nem circular nem linear. Mediante a análise das obras, percebemos que Cícero retrata a República de seu tempo, ou seja, do presente, como decadente, e apenas haveria expectativa de melhora se as ações exemplares do passado fossem recuperadas. / The aim of the thesis is to analyze the relation between politics and history in Ciceronian work. Initially, we examine the conception of man, especially the wise-politician, and his capacity for action in the republic. We analyze human action portrayed in philosophical dialogues, discourses and historical narratives; we observe that political action is the subject of historical narratives and historical examples are constitutive of political-philosophical argumentation. Thus, political works are particularized by historical examples and historical narratives are universalized by the presence of political-philosophical thought. According to the precept of historical narrative, resorting to history has in all works a pedagogical-political function, preserving the actions worthy of memory of the past, speaking to the present time and being able to be extended to the future. Through the analysis of the use of history we observe a Ciceronian conception of the course of history in Rome, which is neither circular nor linear. Through the analysis of the works we understand that Cicero describes the Republic of his time, that is, of the present, as decadent, and there would be only expectation of improvement if the exemplary actions of the past were recovered.

The thesis is a defence of an original position in the philosophy of action. It argues for a pluralist view of actions dubbed Strong Pluralism. One of the key questions of philosophy of action since the early 20th century has been taken to be 'What are actions'? In my thesis I argue that there is no single correct answer to this question. I put forward two positive claims which explain why this is so: 1. That 'action' is ambiguous and can mean either doing or thing done. 2. That not all doings fall into the same metaphysical category because they can have different constitutive structures: some of them are causings, some are events, and others are processes. I demonstrate in the thesis that these two claims can be held coherently, and I identify the resulting view as Strong Pluralism about action. The thesis divides into two parts. In the first part I lay out and offer a defence of the view in question and in the second I discuss how my pluralist view of relates to the three major types of views of action: events, causings, and process views. The first part of the thesis consists of three steps of the main argument of my thesis, each step outlined and argued for in a chapter. In the first chapter I offer an overview of the answers provided to the 'What is action'? question offered by philosophers in the last 80 years. I identify a trend common to these views to advance monist answers, that is, they offer views of action which are committed to 'action' meaning one thing and all actions fitting into the same metaphysical category. I argue that the monist answers are unsatisfactory and monism about actions cannot be maintained. In chapter two I offer an alternative to monism in the form of pluralism about actions. I defend pluralism by arguing that 'action' is ambiguous between doing and things done, and by showing that it is a as suitable substitute for monism. I provide an overview of the four most important ways in which the doing/thing done distinction has been made, and I suggest and defend a further version of it. In chapter three, I outline three possible pluralist views of actions, and defend the view which I call Strong Pluralism. Strong Pluralism is committed to the claims that there are both doings and things done, and that there are doings which have different constitutions from other doings, hence it is correct to think that some doings are events, some are causings, and some are processes. In the three chapters which constitute the second part of the thesis I engage successively with views which have claimed that actions are events, that they are causings, and that actions are processes. I argue in each chapter that there are doings which can be said to belong into the category discussed and I provide positive accounts and examples of when this is so. I offer a categorization of doings which helps us to decide which doings fall into the group of events, which into that of causings, and which into that of processes. Throughout these three chapters I critically discuss the most influential events, causings, and process views and point out several aspects in which they are too limiting or mistaken about doings. The view of doings and things done worked out in thesis helps to resolve long standing issues in the philosophy of action by clarifying what we take to be the object of explanation, knowledge, and evaluation when we discuss actions in ethics, philosophy of science, philosophy of mind, epistemology (esp. debates about knowledge of action) and other fields of philosophy. The view can have broader applications in the fields of moral psychology and cognitive science by helping to sharpen our account of what researchers are discussing when they are discussing actions.

Automation is fundamentally changing many industries and retail is no exception. Moonshopis a South African venture trying to solve the problem of autonomous grocery storesusing cameras and computer vision. This project is the continuation of a hackathon heldto explore different methods for Human Action Recognition in Moonshop’s stores.Throughout the project a pipeline for data processing has been developed and two typesof Graph-Convolutional Networks, CTR-GCN and ST-GCN, have been implementedand evaluated on the data produced by this pipeline. The resulting scores aren’t goodenough to call it a success. However, this is not necessarily a fault of the models. Rather,there wasn’t enough data to train on and the existing data was of varying to low quality.This makes it complicated to justly judge the models’ performances. In the future, moreresources should be spent on generating more and better data in order to really evaluatethe feasibility of using Human Action Recognition and Graph-Convolutional Networksat Moonshop.

No description available.

Static image action recognition aims at identifying the action performed in a given image. Most existing static image action recognition approaches use high-level cues present in the image such as objects, object human interaction, or human pose to better capture the action performed. Unlike images, videos have temporal information that greatly improves action recognition by resolving potential ambiguity. We propose to leverage a large amount of readily available unlabeled videos to transfer the temporal information from video domain to static image domain and hence improve static image action recognition. Specifically, We propose a video prediction model to predict the future video of a static image and use the future predicted video to improve static image action recognition. Our experimental results on four datasets validate that the idea of transferring the temporal information from videos to static images is promising, and can enhance static image action recognition performance. / Master of Science

En la actualidad, la cámara de vídeo se ha convertido en un dispositivo omnipresente. Debido a su miniaturización, estas se pueden encontrar integradas en multitud de dispositivos de uso diario, desde teléfonos móviles o tabletas, hasta ordenadores portátiles. Aunque estos dispositivos son empleados por millones de personas diariamente de forma inofensiva, capturando vídeo, realizando fotografías que luego son compartidas, etc.; el empleo de videocámaras para tareas de videovigilancia levanta cierta preocupación entre la población, sobre todo cuando estas forman parte de sistemas inteligentes de monitorización. Esto supone una amenaza para la privacidad debido a que las grabaciones realizadas por estos sistemas contienen una gran cantidad de información que puede ser extraída de forma automática mediante técnicas de visión artificial. Sin embargo, la aplicación de esta tecnología en diversas áreas puede suponer un impacto muy positivo para las personas. Por otro lado, la población mundial está envejeciendo rápidamente. Este cambio demográfico provocará que un mayor número de personas en situación de dependencia, o que requieren apoyo en su vida diaria, vivan solas. Por lo que se hace necesario encontrar una solución que permita extender su autonomía. La vida asistida por el entorno (AAL por sus siglas en inglés) ofrece una solución aportando inteligencia al entorno donde residen la personas de modo que este les asista en sus actividades diarias. Estos entornos requieren la instalación de sensores para la captura de datos. La utilización de videocámaras, con la riqueza en los datos que ofrecen, en entornos privados haría posible la creación de servicios AAL orientados hacia el cuidado de las personas como, por ejemplo, la detección de accidentes en el hogar, detección temprana de problemas cognitivos y muchos otros. Sin embargo, dada la sencilla interpretación de imágenes por las personas, esto plantea problemas éticos que afectan a la privacidad. En este trabajo se propone una solución para poder hacer uso de videocámaras en entornos privados con el objetivo de dar soporte a las personas y habilitar así el desarrollo de servicios de la vida asistida por el entorno en un hogar inteligente. En concreto, se propone la protección de la privacidad en aquellos servicios AAL de monitorización que requieren acceso al vídeo por parte de un cuidador, ya sea profesional o informal. Esto sucede, por ejemplo, cuando se detecta un accidente en un sistema de monitorización y ese evento requiere la confirmación visual de lo ocurrido. Igualmente, en servicios AAL de telerehabilitación puede ser requerida la supervisión por parte de un humano. En este tipo de escenarios es fundamental proteger la privacidad en el momento en que se esté accediendo u observando el vídeo. Como parte de este trabajo se ha llevado a cabo el estudio del estado de la cuestión en la cual se han revisado los métodos de protección de la privacidad visual presentes en la literatura. Esta revisión es la primera en realizar un análisis exhaustivo de este tema centrándose, principalmente, en los métodos de protección. Como resultado, se ha desarrollado un esquema de protección de la privacidad visual basado en el reconocimiento del contexto que permite adecuar el nivel de privacidad durante la observación cuando las preferencias del usuario coinciden con el contexto. La detección del contexto es necesaria para poder detectar en la escena las circunstancias en que el usuario demanda determinado nivel de privacidad. Mediante la utilización de este esquema, cada uno de los fotogramas que componen un flujo de vídeo en directo es modificado antes de su transmisión teniendo en cuenta los requisitos de privacidad del usuario. El esquema propuesto hace uso de diversas técnicas de modificación de imágenes para proteger la privacidad, así como de visión artificial para reconocer dicho contexto. Por tanto, en esta tesis doctoral se realizan diversas contribuciones en distintas áreas con el objetivo de llevar a cabo el desarrollo del esquema propuesto de protección de la privacidad visual. De este modo, se espera que los resultados obtenidos nos sitúen un paso más cerca de la utilización de videocámaras en entornos privados, incrementando su aceptación y haciendo posible la implantación de servicios AAL basados en visión artificial que permitan aumentar la autonomía de las personas en situación de dependencia.