A carregar...
Projeto de investigação
Sem título
Financiador
Autores
Publicações
A computational approach to the art of visual storytelling
Publication . Marcelino, Gonçalo Barreto Ferreira; Magalhães, João
For millennia, humanity as been using images to tell stories. In modern society, these
visual narratives take the center stage in many different contexts, from illustrated children’s
books to news media and comic books. They leverage the power of compounding
various images in sequence to present compelling and informative narratives, in an immediate
and impactful manner. In order to create them, many criteria are taken into account,
from the quality of the individual images to how they synergize with one another.
With the rise of the Internet, visual content with which to create these visual storylines
is now in abundance. In areas such as news media, where visual storylines are regularly
used to depict news stories, this has both advantages and disadvantages. Although content
might be available online to create a visual storyline, filtering the massive amounts
of existing images for high quality, relevant ones is a hard and time consuming task. Furthermore,
combining these images into visually and semantically cohesive narratives is a
highly skillful process and one that takes time.
As a first step to help solve this problem, this thesis brings state of the art computational
methodologies to the age old tradition of creating visual storylines. Leveraging
these methodologies, we define a three part architecture to help with the creation of visual
storylines in the context of news media, using social media content. To ensure the
quality of the storylines from a human perception point of view, we deploy methods for
filtering and raking images according to news quality standards, we resort to multimedia
retrieval techniques to find relevant content and we propose a machine learning based
approach to organize visual content into cohesive and appealing visual narratives.
BioVisualSpeech: uma plataforma com jogos sérios de apoio à terapia da fala que permite parametrização dos treinos
Publication . Martins, Sofia Trigó Rodrigues; Cavavo, Sofia
A fala permite a transmissão de diversos tipos de informação importantes, sendo o
meio de comunicação mais utilizado e, por isso, um elemento essencial para a integração
na comunidade constituindo, deste modo, um dos aspetos mais importantes na nossa
vida. Infelizmente, os desvios da fala são um problema comum entre muitas crianças e,
se não forem tratados, podem vir a causar um impacto direto não só na vida social da
criança como também na sua vida académica. Assim sendo, a terapia da fala tem um
papel muito importante uma vez que é através dela que as crianças podem superar os
referidos desvios.
A terapia da fala é uma área da saúde que tem vindo a crescer significativamente nos
últimos anos. Sendo uma terapia relativamente recente [3], os recursos que apoiam esta
terapia são escassos tendo de ser, muitas vezes, adaptados e criados pelos terapeutas. Para
além disso, os materiais disponíveis para as sessões de terapia da fala são muito limitados
e monótonos sendo, deste modo, difícil manter as crianças motivadas para a execução dos
exercícios.
Uma forma de motivar as crianças é através de jogos eletrónicos, tecnologias com que,
hoje em dia, elas já se encontram bastante familiarizadas. Por conseguinte, propõe-se uma
plataforma de software, através de técnicas de gamificação da terapia, que compreendam
um sistema de recompensas, dificuldades adaptadas consoante as necessidades de cada
criança, análise de som e ainda a possível parametrização por parte dos terapeutas. Esta
plataforma de software inclui dois jogos distintos, permitindo, contudo, a posterior adição
de outros jogos.
Ambos os jogos utilizam um sistema automático de reconhecimento de fala de modo
a classificar as produções de fala das crianças. São ambos dependentes do que a criança
diz e oferecem feedback visual imediato sobre o mesmo. Têm ainda algumas pistas para
auxiliar as crianças, caso as mesmas apresentem alguma dificuldade, e um sistema de re compensas de modo a manter a sua motivação. Um dos jogos contém ainda diversos níveis
e um sistema de dificuldade automático. Estes jogos pretendem assim, auxiliar as crianças
na superação das suas dificuldades com um determinado género de sons, podendo ser
utilizados tanto em sessões de terapia, como em casa para as crianças treinarem. O(a)
terapeuta terá a facilidade em seguir a criança de modo a possibilitar uma articulação das sessões de terapia mediante as facilidades/dificuldades apresentadas pelas crianças.
Temporal Information Models for Real-Time Microblog Search
Publication . Martins, Flávio Nuno Fernandes; Magalhães, João; Callan, Jamie
Real-time search in Twitter and other social media services is often biased
towards the most recent results due to the “in the moment” nature of topic
trends and their ephemeral relevance to users and media in general. However,
“in the moment”, it is often difficult to look at all emerging topics and single-out
the important ones from the rest of the social media chatter. This thesis proposes
to leverage on external sources to estimate the duration and burstiness of live
Twitter topics. It extends preliminary research where itwas shown that temporal
re-ranking using external sources could indeed improve the accuracy of results.
To further explore this topic we pursued three significant novel approaches: (1)
multi-source information analysis that explores behavioral dynamics of users,
such as Wikipedia live edits and page view streams, to detect topic trends
and estimate the topic interest over time; (2) efficient methods for federated
query expansion towards the improvement of query meaning; and (3) exploiting
multiple sources towards the detection of temporal query intent. It differs from
past approaches in the sense that it will work over real-time queries, leveraging
on live user-generated content. This approach contrasts with previous methods
that require an offline preprocessing step.
Musiquence – Design, Implementation and Validation of a Customizable Music and Reminiscence Cognitive Stimulation Platform for People with Dementia
Publication . Ferreira, Luís Duarte Andrade; Badia, Sergi; Cavaco, Sofia
Dementia is a neurodegenerative disease that affects millions of individuals worldwide and is challenging to diagnose as symptoms may only perceivable decades later. The disease leads to a gradual loss of memory, learning, orientation, language, and comprehension skills, which compromises activities of daily living. Health-related costs caused by dementia will continue to increase over the next few years; between the years 2005 and 2009, an increase of 34% (from $315 to $422 billion worldwide) was observed in treating dementia-related issues. Pharmaceutical approaches have been developed to treat dementia symptoms; unfortunately, the risk of side effects is high. For this reason, nonpharmaceutical methods such as music and reminiscence therapies have gained acceptance as patients with dementia positively respond to such approaches even at later stages of the disease. Nevertheless, further research is needed to understand how music and reminiscence therapy should be used and to quantify their impact on individuals with dementia. The development of serious games has gained attention as an alternative approach to stimulate patients. However, the clinical impact that serious games have on individuals with dementia is still unclear. In this dissertation, we contribute with new knowledge regarding the usage of music and reminiscence approaches in people with dementia through a theoretical model. Based on Baddeley’s working memory model, our model aims to explain how the therapeutic properties of music and reminiscence can have a beneficial effect. To test our model, we developed a novel interactive platform called Musiquence, in which healthcare professionals can create music and reminiscence based cognitive activities to stimulate people with dementia. In this dissertation, we present the results from several studies about the usage and effects that music and reminiscence have on such a population. We performed two studies using Musiquence to study the feasibility of a novel learning method based on musical feedback to aid people with dementia during task performance in virtual reality settings. Results show that participants relied more on music-based feedback during the task performance of virtual reality activities than in other forms of feedback. Also, data suggest that the music-based feedback system can improve task performance, compensating for some dementia-related deficits. We also used Musiquence in a longitudinal one-month-long pilot study to assess its efficacy when used for a cognitive stimulation intervention in dementia patients. The results of the study are promising. The 3 participants showed improvements in terms of general cognition, quality of life, mood, and verbal fluency.
Unidades organizacionais
Descrição
Palavras-chave
Contribuidores
Financiadores
Entidade financiadora
Fundação para a Ciência e a Tecnologia
Programa de financiamento
5665-PICT
Número da atribuição
CMUP-ERI/TIC/0033/2014