Brazil

Audio Description (AD) or Video Description is a vital accessibility concept in blind and visually impaired people&#39;s life. Automating this task is not easy and involves many problems, such as describing the scenario, actions, emotions, and characters. This paper presents an approach to automatically describe characters — in a video or image — combining Deep Learning (DL), Face detection, Facial Expression detection techniques, and audio synthesizers. Our proposal uses the detection tools, applies some DL models to the analyzed data, and generates an audio description. To evaluate the feasibility of our proposal, we have developed a proof of concept of the solution and performed some computational experiments to evaluate it.

WebMedia 2021

An Approach for Automatic Description of Characters for Blind People

audio description

blind people

accessibility

deep learning

### Welcome to the Brazilian Symposium on Multimedia Systems and Web 
Promoted annually by the Brazilian Computer Society (SBC), the Brazilian Symposium on Multimedia Systems and Web (WebMedia) is the main event on the subject in Brazil and an excellent opportunity for scientific and technical exchanges between students, researchers and professionals in the areas of Multimedia, Hypermedia and Web. This year, we will be celebrating the 27th edition of WebMedia, to be held from **November 5th to 12th, 2021**, organized by the Federal University of Minas Gerais and Federal University of São João Del Rei. For the second time in its history, the event will be held online, with interaction between the participants being done through tools and web platforms.
 <iframe width="560" height="315" src="https://www.youtube.com/embed/2GvWKRUDtAY" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> 
The Webmedia 2021 schedule will feature:
 
* Technical sessions for the presentation of scientific articles.
* National and international lectures
* Panel with Companies
* Workshops
* Short courses
* Project meetings open to the community
 

The topics of interest for paper submission are:
 
* Web and Social Networks
* Multimedia
* Digital TV, Ubiquitous and Mobile Computing
 

In addition to the main track, WebMedia 2021 will feature the Tools and Applications Workshop, the Scientific Initiation Papers Contest, the Theses and Dissertations Contest, the RNP CT-Video Workshop.
 
The organization of WebMedia 2021 is under the responsibility of the [Department of Computer Science at UFMG (DCC/UFMG)](http://dcc.ufmg.br/) and [Department of Computer Science at UFSJ (DCOMP/UFSJ)](https://dcomp.ufsj.edu.br/)
 
**#PartiuWebMedia2021 #WebMedia2021**

The Brazilian Symposium on Multimedia Systems and Web (WebMedia) is the main event on the subject in Brazil and an excellent opportunity for scientific and technical exchanges between students, researchers and professionals in the areas of Multimedia, Hypermedia and Web.

Audio Description (AD) or Video Description is a vital accessibility concept in blind and visually impaired people's life. Automating this task is not easy and involves many problems, such as describing the scenario, actions, emotions, and characters. This paper presents an approach to automatically describe characters — in a video or image — combining Deep Learning (DL), Face detection, Facial Expression detection techniques, and audio synthesizers. Our proposal uses the detection tools, applies some DL models to the analyzed data, and generates an audio description. To evaluate the feasibility of our proposal, we have developed a proof of concept of the solution and performed some computational experiments to evaluate it.

technical paper

Online user reviews written in Portuguese language are widely used by native speakers from several countries for decision-making and a large amount user reviews are available on the Web. However, only few studies on Portuguese sentiment analysis are reported as there is a lack of resources including domain-specific sentiment lexical collections. In this paper, we present an effective method for building a set of lexicons in restaurant domain for Portuguese language. The experiments on a large data collection of user reviews illustrate that our method significantly outperforms several alternative previous approaches of building domain-specific sentiment lexicon.

Building a Restaurant-Specific Sentiment Lexicon via Probability Theory

The search for information is inherent to the human being and with the popularization of computers, search systems have become the main source of obtaining knowledge. However, they do not support greater user interaction when they want to do a complex search. It is important to evaluate the user and infer information about him, to know his intention to search to increase the performance of these systems under the principles of SaL. In this paper, we propose a model for classifying the search intention using Transformers. It was used an available dataset and we obtained 97% accuracy in the validation set and 99% f-measure in the test set, achieving better results when comparing to traditional methods.

Predicting the User Intention in Web Search

Professional fact-checking tends to be costly and scalability challenging. For this reason, a series of methods to automate this process has been emerging. Unfortunately, these methods have been created with monolithic architectures, not making use of pre-built parts, and are harder to be understood by others. This work aims to propose a framework, where the development of methods can be done in a modular manner, creating workflows based on key steps that can be linked together to generate the input classification. For this, a study of the literature was conducted to identify the main processing steps of the methods, seeking to develop a solution that would guarantee flexibility of execution and interoperability between the different processing components. With this, the automatic verification process was described in a limited set of steps that can be performed, creating independent components that implement each of them. The framework makes use of a proposed data traceability ontology to map all generated data during execution under a unified vocabulary, which facilitates communication between these independent components.

ReVera Framework: A Framework for Fact Checking Traceability

Exploring large news collections created by media outlets with traditional search engines is impractical for demanding users. We propose a temporal exploration tool that aims to facilitate the consultation of news collections. We concentrated our efforts on two fronts: (i) allowing users to make queries with the addition of information from documents represented by word embeddings, and; (ii) retrieving temporal information to generate timelines presented by an appropriate interface. We evaluated our solution in a collection of a Brazilian newspaper, demonstrating that it can draw different timelines, covering different subtopics of the same theme.

Semantically Time Tracking of Events from Web Documents

Anaglyph Stereo Matching is a particular, harder, case of the widelystudied Stereo Matching algorithms. When applied to AnaglyphReversion problem helps to improve stereo content storage andtransmission, stereoscopic visualization independence and accessto legacy anaglyph content available on the Web. However, thereversion process is not trivial since anaglyphs have high radiome-tric differences between the two views of a stereo pair of images,preventing standard stereo matching algorithms to be directly used.Different from related work, this paper presents a new pixel descrip-tor specially designed to deal with radiometric differences, allowingfor the improvement of the final reconstructed stereo pair imagequality. Experimental results show the proposed method presents,in average, better image quality when compared with related tech-niques.

PixRadio - a new anaglyphic reversion technique based on radiometric differences reduction

Adding sensory effects to applications has proven to increase users’
quality of experience. As immersive technologies advance in the fields of entertainment, education, training, simulation, and so forth,
more fine-grained functions are craved by users to have truly captivating and (nearly) realistic experiences. These new functions, in
turn, push forward both the industry and academia on developing
solutions to overcome the current limitations. This is indeed the
particular case of multimedia systems enriched with thermal wind
sensations. Whilst both wind and thermal effects have been included separately in such systems, the lack of wind at a low or high
temperature has been noticed by users. It is not trivial building,
controlling, and working with devices at different temperatures hit
by wind at the same time. This is exacerbated by the fact that different technologies are used to this end. In this paper, we introduce
a haptic system that allows the creation and delivery of hot and
cold airflow synchronised with mulsemedia content. The system
combines the vaporisation of cooled water by Peltier elements for
cold wind and hair dryers for warm wind steered by wind fans. To
assess our system, we analysed the time our devices take to deliver
these effects considering different starting temperatures in the environment. Furthermore, we annotated three videos with hot wind,
cold wind, and both effects respectively to observe the feasibility
of having those effects integrated with audiovisual content. Our
findings suggest recommendations for the appropriate triggering
of these types of sensory effects combined with multimedia content
taking into account our setup.

A Haptic System for Switching Wind Temperatures Based on Ultrasonic Vibrations, Peltier Elements, and Electrical Resistances for Multisensory Applications

Multisensory books and audiobooks provide users with greater immersion by stimulating multiple human senses simultaneously during the reading process. A multisensory story contains annotations of effects synchronizing each sensorial stimulus to traditional media, and the manual authoring of these annotations is a costly and time-consuming task. While there are authoring tools to synchronize multisensory effects with videos, there are no authoring tools available for multisensory audiobooks, to the best of our knowledge. This article proposes an authoring tool for multisensory audiobooks called MAAT. It presents the evaluation of MAAT´s QoE aspects and compatibility assessments on the file generated by the MAAT tool and on the synchronization of multisensory effects.

MAAT: Multisensorial Audiobooks Authoring Tool

The presentation of sensory effects in sync with 360◦ content has the potential to increase user immersion. However, the authoring process for such effects is laborious and slow. his paper brings two contributions to address this need: a generic architecture for inferring the presentation characteristics of sensory effects in 360◦content and the implementation of a neural network capable of locating objects related to sensory effects in equirectangular images. The proposed architecture decouples and eases the adaptation of multimedia content analysis techniques for recognizing sensory effects. In this paper, the proposed architecture is instantiated to use two inference modules.

Sensory Effect Extraction for 360 Media Content

Sign languages (LS) are the natural form of communication between
deaf people. However, although sign languages are recognized in
many countries as a language, access to information and education
is still a common problem for deaf people. In Brazil, despite the
existence of several accessibility laws that determine support for
deaf people, many are still dependent on Libras interpreters to
communicate, which impairs their autonomy. As a result, there
is a great difficulty for the deaf to exercise their citizenship and
have access to quality information and education through sign
languages, which generally implies linguistic delay and knowledge
acquisition. In this sense, the present work addresses a proposal for
an accessible interface of tele-education for the hearing impaired
in order to facilitate access to information, making it possible to
bring the figure of the remote interpreter to the classroom. This
solution involves adapting the interface of the open source system,
making it possible for the remote interpreter to be present to meet
the needs of deaf students and make the classroom experience
more accessible. In addition, a solution for displaying live subtitles
was also implemented. The results obtained in this study proved
to be promising, with 50% of users successfully completing more
than 93.33% of the requested tasks. In addition, the group with the
proposed solution was less dispersed when compared to the other
group, concentrating most of its results above 80%, in a way that
leads to believe that this type of adaptation can be welcomed by
deaf users in the context of tele-education.

Uma proposta de interface de videoconferência acessível para pessoas com deficiência auditiva

Cardiovascular diseases (CVDs) are the leading cause of death worldwide. Approximately 1.13 billion people suffer from arterial hypertension (AH), a condition that stands out among all other CVD risk factors. One of the main non-pharmacological treatments for AH is the performance of physical activities. Studies have also demonstrated the possibility of blood pressure control through the application of breathing exercises. Such exercises can be performed even at home. Due to the limitations concerning physical exercise for patients with uncontrolled hypertension, clinicians often have their patients conduct breathing exercises before a physiotherapy session for blood pressure reduction purposes. However, more tools are needed to assist clinicians with this approach. Considering these circumstances, this paper introduces PhysioAR, an augmented reality system designed to assist in respiratory therapy treatments for hypertensive patients.
Using three-dimensional virtual environments and specific hardware for vital sign measurements, PhysioAR enables the creation and application of respiratory therapy sessions by a physiotherapist professional. At the end of its development, a group of usability experts evaluated the system through a questionnaire based on Nielsen’s heuristic evaluation of user interfaces.
After evaluation, PhysioAR obtained an average performance of 4.3 by the Likert scale. All evaluators agreed that using the system was a pleasant experience, which suggests that PhysioAR has potential for physiotherapy clinics. The questionnaire application exposed some user experience problems. This demonstrated the effectiveness of a questionnaire-based evaluation to identify possible system problems.

An Approach for Automatic Description of Characters for Blind People

Downloads

Next from WebMedia 2021

Building a Restaurant-Specific Sentiment Lexicon via Probability Theory

Similar lecture

Uma proposta de interface de videoconferência acessível para pessoas com deficiência auditiva

An Approach for Automatic Description of Characters for Blind People

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from WebMedia 2021

Building a Restaurant-Specific Sentiment Lexicon via Probability Theory

Similar lecture

Uma proposta de interface de videoconferência acessível para pessoas com deficiência auditiva

Downloads