Brazil

Este trabalho propõem uma abordagem para auxiliar pessoas com deficiência visual no reconhecimento de pessoas independente da idade. O objetivo é desenvolver um sistema que utilize uma abordagem de reconhecimento facial, com foco na invariância na idade, que retorne bons resultados comparados aos resultados obtidos na revisão da literatura. A abordagem estudada utiliza Redes Neurais Convolucionais profundas - CCNs, pré-treinadas pelo conjunto de dados VGGFace2, para extrair descritores de características de imagens de faces e classificar com o algoritmo de classificação Linear SVM. Como pode ser visto no decorrer do trabalho, a abordagem retornou 89,9% de acurácia, utilizando o conjunto de dados FG-NET, com 1002 imagens. E utilizando o conjunto de dados CACD, que contém 163.446 imagens divididas em quatro subconjuntos diferentes, três conjuntos para treino e um para teste, a abordagem retornou 85,2%, 82,4% e 88,2% de acurácia para cada modelo treinado com um conjunto de treinamento diferente.

BRACIS 2021

Age-Invariant Face Recognition to Aid Visually Impaired Pleople

pessoas com deficiência visual

reconhecimento facial

redes neurais

### 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**

WebMedia 2021

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.

technical paper

**BRACIS** originated from the combination of the two most important scientific events in Brazil in Artificial Intelligence (AI) and Computational Intelligence (CI): the Brazilian Symposium on Artificial Intelligence – **SBIA** (22 editions), and the Brazilian Symposium on Neural Networks – **SBRN** (13 editions). **ENIAC** is a national forum for researchers, practitioners, educators and students to present and discuss innovations, trends, experiences, developments and work in progress in the fields of Artificial and Computational Intelligence. **STIL** is the bi-annual Language Technology event.
All events are supported by the Brazilian Computer Society – Special Interest Groups: **CE-IA**, **CE-IC** and **CE-PLN**.

To access the 10th Brazilian Conference on Intelligent Systems (**BRACIS**), you need to register for the conference.  Please visit the Conference official site to register

[Register now](http://c4ai.inova.usp.br/bracis2021/)

Register now

BRACIS originated from the combination of the two most important scientific events in Brazil in Artificial Intelligence (AI) and Computational Intelligence (CI): the Brazilian Symposium on Artificial Intelligence – SBIA (22 editions), and the Brazilian Symposium on Neural Networks – SBRN (13 editions). 
ENIAC is a national forum for researchers, practitioners, educators and students to present and discuss innovations, trends, experiences, developments and work in progress in the fields of Artificial and Computational Intelligence. 
STIL is the bi-annual Language Technology event.
All events are supported by the Brazilian Computer Society – Special Interest Groups: CE-IA, CE-IC and CE-PLN

Policy Search (PS) methods have been used to learn optimization algorithms, obtaining encouraging results. In this work, we used PS methods to learn optimization algorithms for global optimization problems, considering a little-studied scenario: high-dimensional functions and the optimization algorithms do not have access to the derivatives of the function to be optimized. The results indicate that, despite the difficulties, the learned optimization algorithms have promising performance in the studied scenario.

Zeroth Order Policy Search Methods for Global Optimization Problems: An Experimental Study

There is a growing interest in systems that detect violence automati-
cally using ambient audio. In this work, we built and evaluated four classifiers
with this proposal. However, instead of directly processing the audio signals, we
converted them to images, known as Mel-spectrograms, and then used Convolu-
tional Neural Networks (CNN) to treat as an image classification problem using
pre-trained networks in this context. We tested the architectures Inception v3,
VGG-16, MobileNet v2, and ResNet152 v2, and the classifier coming from the
MobileNet architecture obtained the best classification results when evaluated
on the HEAR Dataset created to carry out this research.

Deep Learning and Mel-spectrograms for Physical Violence Detection in Audio

We propose a solution based on Multi-Layer Perceptron (MLP) to predict the offset of the center of gravity of an offshore platform. We also perform a comparative study with three optimization algorithms – Random Search, Simulated Annealing, and Bayesian Optimization (BO) – to find the best MLP architecture. Although BO obtained the best architecture in the shortest time, ablation studies developed with hyperparameters of the optimization process showed that the result is sensitive to them and deserves attention in the Neural Architecture Search process.

Evaluation of Neural Architecture Search Approaches for Offshore Platform Offset Prediction

Training deep neural networks is a relevant problem with open questions related to convergence and quality of learned representations. Gradient-based optimization methods are used in practice, but cases of failure and success are still to be investigated. In this context, we set out to better understand the convergence properties of different optimization strategies, under different parameter options. Our results show that (i) feature embeddings are impacted by different optimization settings, (ii) suboptimal results are achieved by the use of default parameters, (iii) significant improvement is obtained by making educated choices of parameters, (iv) learning rate decay should always be considered. Such findings offer guidelines for training and deployment of deep networks.

Optimization Matters: Guidelines to Improve Representation Learning with Deep Networks

Automatic gender recognition from speech is a problem related to the
area of speech analysis and has a variety of applications thats extends from the
personalisation of product recommendation to forensics. Identifying the efficiency and costs of different approaches that deal with this problem is imperative. This work aims to investigate and compare the efficiency and costs of different deep learning architectures in the task of gender recognition from speech.

A Comparison of Deep Learning Architectures for Automatic Gender Recognition from Audio Signals

Algorithms for learning optimal paths are present in several scenarios. Given this, LRTA* (learning real time A*) appears as an option for reconciling planning and action. This article investigate how the variation of the number of agents impacts on the distances traveled by them to find optimal path using the LRTA* in static environments. Experiments show the existence of a relationship that the grater number of agents, the total quantity and per capita movements behave like mathematical curves, the first one like a linear and the second as a decreasing exponential. Through the relationship, it is possible to define the best number of agents in the search for the optimal path in terms of performance.

Impact of the variation of the number of agents in the cooperative learning of optimal paths using LRTA*

Electronic Legal Proceedings are worldwide legal phenomena, allowing
the use of computerized systems for the creation and monitoring of procedural
acts in the most diverse legal bodies. On one hand, it allows greater
transparency in the conduct of procedural acts, on the other, it has contributed
to the bottleneck of open but unresolved lawsuits each year. Nowadays, Information
Retrieval to automate the processing of these procedural objects is at the
forefront of computer systems for Law. In this study, we present MISLA2, a system
to retrieve orders and preliminaries from judicial labor sentences through
ontological models built from previous cases. Instead of tied and difficult-to-maintain
domain specification models, we demonstrate how light ontologies, in
conjunction with regular expressions for extracting significant portions of the
text, can achieve the desired results. In addition, empirical experiments carried
out with real labor lawsuits evidence that results are quite promising.

MISLA2: A System to Information Retrieval in Labor Lawsuits using Legal Ontologies and Regular Expressions

In a multi-agent system (MAS), it is very usual that agents delegate tasks to each other. However, due to the subjectivity of the information used by agents during the decision-making process, an agent may end up delegating a task to an untrustworthy partner. In this work, we present a trust computing approach based on quantitative argumentation with votes (\hbox{QuAD-V}), where a trust measure is estimated according with the agents' opinions about the service provided by a partner. Besides, such an approach provides a mechanism to evaluate the credibility of agents that play as information sources. In our results, we demonstrate how our trust computing approach can be employed for detecting lying agents, which are able to slander or promote other agents.

Trust Computing Based on Argumentation Debates with Votes for Detecting Lying Agents

A taxa de perfuração (ROP) de um poço de petróleo é uma métrica muito importante de se controlar, por influenciar na produtividade do poço, no desgaste da broca e na segurança do poço e operação. Este trabalho avalia a utilização do algoritmo de floresta aleatória para previsão da ROP em poços de petroleo offshore como alternativa para modelos tradicionalmente utilizados pela indústria. Modelos de floresta aleatória foram avaliados com diferentes hiper-parametros para 4 poços offshore no pré-sal da Bacia de Santos. Não houve uma única combinação de hiper-parâmetros ótima comum entre os poços.

Previsão de Taxa de Perfuração em Poços de Petróleo Offshore Utilizando Algoritmos de Aprendizado de Máquina

The core point of the research process are data. They are records fromscientific investigation, which support the results published in journals and con-ferences. Making research data available in open access digital repositories hasmany advantages, such as increasing the visibility of associated publications,reproducing experiments, and validating results. In Brazil, full and unrestrictedsharing of them is not yet accepted by most researchers. This paper presents aninitial study to describe a model analyzing the attitude of Brazilian researchersconcerning open access research data. A clustering algorithm was used to iden-tify different research profiles. The achieved results indicate the main reasonswhy the researchers object to share their data

Premium content

Age-Invariant Face Recognition to Aid Visually Impaired Pleople

Downloads

Next from BRACIS 2021

Zeroth Order Policy Search Methods for Global Optimization Problems: An Experimental Study

Similar lecture

Quality Enhancement of Highly Degraded Music Using Deep Learning-Based Prediction Models for Lost Frequencies