Este trabalho recorre a técnicas de Visão por Computador e de Computação Gráfica para caracterizar o movimento 3D de corpos articulados. As técnicas de Visão por Computador são utilizadas para efectuar o seguimento de marcas e para obter a sua posição tridimensional em relação às câmaras. Com recurso a técnicas de Computação Gráfica é efectuada uma animação do movimento tridimensional do corpo articulado, a qual poderá ser um auxiliar precioso, por exemplo, no domínio da Biomecânica. No âmbito deste trabalho, foi efectuada uma aplicação ao caso do estudo da estabilidade articular do tornozelo. Os resultados obtidos mostram que a metodologia é viável.

Traditional content creation for computer games is a costly process. In particular, current techniques for authoring destructible behaviour are often limited to a single object basis. In this paper, we build on previous research results to develop a novel method for designing reusable destructible behaviour which can be applied in real-time to a variety of objects. To separate the destructible behaviour from particular objects, we introduce the concept of destructible materials: where the material of an object usually defines the way an object looks, a destructible material will determine how it breaks. Destructible materials provide a reusable definition and intuitive way of designing and tweaking destructible behaviour of objects in game development, which can then be applied in real-time.

In this work we propose an integration-aware super-resolution approach for 3D vector fields. Recent work in flow field superresolution has achieved remarkable success using deep learning approaches. However, existing approaches fail to account for how vector fields are used in practice, once an upsampled vector field is obtained. Specifically, a cornerstone of flow visualization is the visual analysis of streamlines, or integral curves of the vector field. To this end, we study how to incorporate streamlines as part of super-resolution in a deep learning context, such that upsampled vector fields are optimized to produce streamlines that resemble the ground truth upon integration. We consider common factors of integration as part of our approach - seeding, streamline length - and how these factors impact the resulting upsampled vector field. To demonstrate the effectiveness of our approach, we evaluate our model both quantitatively and qualitatively on different flow field datasets and compare our method against state of the art techniques.

Communication is a critical component of human interaction in order to satisfy basic human needs. This is especially true in technology-supported learning environments, where establishing multiple communicational channels between instructors and students is a requisite to enhance learning. Therefore, specific intervention strategies are required to effectively adjust each channel to satisfy students' communication needs while learning. However, not all online communication channels achieve the goal of including students into their learning communities. Thus, designers should understand what works best in facilitating effective communication and learning. The bottom line lies in how to foster online communication and what impact it may have on effective learning. This paper aims at exploring student use of asynchronous and synchronous communication tools, namely fora and chat tools, and their participation within a technology-supported Multimedia course. We present ongoing research results related to this specific issue, together with implications for designing b-learning experiences, lessons learnt and future work. By this we aim at contributing to understand the impact of online communication tools on student behavior and learning results.

Left-ventricle functional analysis is of paramount importance for the assessment of coronary artery disease. Non-invasive coronary artery imaging using multiple-detector computed tomography (MDCT) provides a wealth of data concerning multiple cardiac phases. These have still to be explored in order to assess if further insight is possible regarding regional and functional analysis of the left ventricle leading to a more informed diagnosis and prognosis. Following on previous work regarding left-ventricle segmentation of all available cardiac phases in MDCT exams this article presents several parameters characterizing global and regional left ventricle function, computed for several cardiac phases and represented over time. Synchronized viewing of the different parameters is also supported.

Parametric CAD, in conjunction with 3D-printing, is democratizing design and production pipelines. End-users can easily change parameters of publicly available designs, and 3D-print the customized objects. In research and industry, parametric designs are being used to find optimal, or unique final objects. Unfortunately, for most designs, many combinations of parameter values are invalid. Restricting the parameter space of designs to only the valid configurations is a difficult problem. Most publicly available designs do not contain this information. Using ideas from program analysis, we synthesize constraints on parameters of parametric designs. Some constraints are synthesized statically, by exploiting implicit assumptions of the design process. Several others are inferred by evaluating the design on many different samples, and then constructing and solving hypotheses. Our approach is effective at finding constraints on parameter values for a wide variety of parametric designs, with a very small runtime cost, in the order of seconds.

Em diversas áreas de aplicação, a simplificação das malhas poligonais usadas para representar os modelos, ao diminuir a sua complexidade, constitui uma possível solução para permitir a sua utilização interactiva. Em geral, uma malha obtida após um processo de simplificação possui diferenças em relação à malha original, que podem ser avaliadas usando diferentes índices de qualidade. No entanto, não é ainda claro o modo como essas medidas quantitativas (p.ex., o desvio geométrico) se relacionam com a percepção que os utilizadores têm da qualidade dos modelos simplificados. Neste artigo, com base num conjunto de modelos pulmonares definidos por malhas triangulares e para três métodos de simplificação particulares, apresentam-se os resultados da avaliação da qualidade dos modelos simplificados usando dois processos distintos: o cálculo de índices de qualidade e a realização de uma experiência controlada com observadores. Os dados obtidos pelos dois processos são comparados, sendo possível estabelecer uma relação entre ambos. Verifica-se que, nestas condições específicas, alguns dos índices quantitativos usados podem servir como estimadores da qualidade das malhas percebida pelos utilizadores.

!-

The presence or not of collisions between objects is usually required to study the interaction between them, increasing the realism in virtual environments. Collision detection between polygonal objects has been widely studied, and more recently some studies have been made concerning collisions between volume objects. Collision detection between volume datasets and polygonal objects is introduced in this work. This kind of mixed scenes appears naturally in many applications such as surgery simulation and volume edition. To detect the collision, first the volume dataset is represented by a single 3D texture. Then, a mapping from eye space to volume space is established, such as each mesh fragment has a 3D texture coordinate. The collision is verified by fragment during the rasterization stage. We use OpenGL R occlusion query extension to count the number of mesh fragments colliding with the volume. Our tests show that up to 3800 pairs of volume-mesh may be evaluated in one second.

Durante a concepção de um molde, os desenhadores consultam regularmente projectos existentes de modo a reaproveitar soluções anteriormente desenvolvidas. Desta forma conseguem reduzir o tempo de desenvolvimento de um molde. No entanto, as soluções existentes para encontrar o que desejam em desenhos anteriores são lentas e complexas. Com o intuito de melhorar o processo de pesquisa de desenhos, desenvolvemos uma abordagem capaz de efectuar a indexação automática de desenhos de moldes e que permite a sua recuperação através esboços, imagens ou uma combinação de ambos. Neste artigo apresentamos essa abordagem, bem como os respectivos protótipos e a avaliação experimental a que estes foram sujeitos. Testes com utilizadores mostraram que o conceito apresentado agradou aos desenhadores e que os nossos algoritmos são simultaneamente e cientes e precisos.