Brazil

A crucial task to overall video understanding is the recognition and localisation in time of different actions or events that are present along the scenes. To address this problem, action segmentation must be achieved. Action segmentation consists of temporally segmenting a video by labeling each frame with a specific action. In this work, we propose a novel action segmentation method that requires no prior video analysis and no annotated data. Our method involves extracting spatio-temporal features from videos using a pre-trained deep network. Data is then transformed using a positional encoder and finally a clustering algorithm is applied with the use of a silhouette score to find the optimal number of clusters where each cluster presumably corresponds to a different single and distinguishable action. In experiments, we show that our method produces competitive results on Breakfast and Inria Instructional Videos dataset benchmarks.

WebMedia 2021

A Cluster-Based Method for Action Segmentation Using Spatio-Temporal and Positional Encoded Embeddings

video segmentation

machine learning

computer vision

The purpose of the AAAI conference is to promote research in artificial intelligence (AI) and scientific exchange among AI researchers, practitioners, scientists, and engineers in affiliated disciplines. 

AAAI 2022

technical paper

### 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.

The advancement of mobile technologies combined with assistive technologies can provide autonomy to people with different types of disabilities, including those with Autistic Spectrum Disorder (ASD). Technologies, such as those found in mobile devices, are very attractive to people with ASD and can be used as a valuable educational tool for these individuals. The aim of this study was to learn about how mobile apps aimed at people with ASD are developed and evaluated, carried out through a systematic mapping of applications aimed at people with ASD, in order to better understand their foundations, motivations, evaluation mode, resources and user profile. The results indicated that the main procedures used to support the applications for people with ASD are the questionnaires to identify the ASD, such as the CHAT and AQ-10, and the ABA intervention, and that these applications, mostly, seek to identify the ASD in a practical way and at a low cost. Most of the apps were evaluated by usability testing and through indirect observation through the app’s logs. The most used hardware and software resources are the mobile device camera and convolutional neural networks (CNN), respectively, with most applications aimed at children and the most common software engineering methodologies were requirements analysis and prototyping.

Systematic mapping of the literature on mobile apps for people with Autistic Spectrum Disorder

Drowsiness is a problem that affects traffic on a worldwide scale, especially among motorcyclists, who use a naturally dangerous means of transport. Nowadays, there is no product on the market that performs an anti-drowsiness action in this public. This article aims to propose the creation of a wearable device coupled to the motorcycle helmet that detects a state of drowsiness through sensors and combats it using actuators. Still at an early stage, a prototype was created, tests in a controlled environment were carried out, preliminary results were presented and a series of future works were drawn up to continue the project.

2DrowSys: an anti-drowsiness device for motorcyclists

Due to the evolution of motion capture devices, natural user interfaces have been applied in several areas, such as neuromotor rehabilitation supported by virtual environments. This paper presents a smartphone application that allows the user to interact with the virtual environment and enables the captured data to be stored, processed, and used in machine learning models. The application submits the recordings to the remote database with information about the movement and in order to apply supervised machine learning. As a proof of concept, we generated a dataset capturing movement data using our application with 232 instances divided into 8 classes of movements. Moreover, we used this dataset for training models that classifies these movements. The remarkable accuracy of the models shows the feasibility of using body articulation data for a classification task after some data transformations.

Upper Limb Motion Tracking and Classification: A Smartphone Approach

Towards a Pragmatic Interoperability on the MIDAS Middleware

The IoT's widely distributed nature has promoted demands for Over-The-Air (OTA) upgrading of embedded systems. This paper presents the conception of a new approach, called iOTA, which meets the security requirements of the SUIT specification of the Internet Engineering Task Force (IETF). The partial evaluation of the proposal points to the continuation of the research and showed that iOTA stands out in terms of security in relation to the other approaches proposed by the literature, offering protection against attacks to which a remote update of an embedded system is usually exposed.

iOTA: An Approach to Secure Over-The-Air Updates on the Internet of Things Scenario

In smart city scenarios, a plethora of data is generated from multiple, decentralized sources, e.g., systems, applications, citizen devices, sensors, etc. However, these data may be not readily available for use due to the high heterogeneity of types, formats, and protocols, thereby giving rise to vertical silos that hamper sharing, integrating, and exploiting these data. This paper introduces Aquedücte, a data integration service for smart cities. Aquedücte offers functionalities for (i) handling data from heterogeneous sources such as files, Web services, systems, etc., (ii) converting these data to a common protocol, and (iii) making them further available to platforms aimed to support the development of smart city applications.

Aquedücte: A Data Integration Service for Smart Cities

Aprendizado de máquina para precaching de conteúdo

Em 2012, 56% das crianças entre 10 e 13 tinham um smartphone, atualmente, uma grande quantidade de dados são compartilhados a cada minuto na internet, inclusive em chats, isso acentua a necessidade de métodos automatizados de moderação de contaúdo. Esse trabalho propôe uma abordagem para controle parental tendo em mente a privacidade dos protegidos enquanto utiliza métodos com deep learning para detectar conteúdo sensível automaticamente.

Should I See or Should I Go?: Automatic Detection of Sensitive Media in Messaging Apps

Fake news can rapidly spread through internet users. Approaches proposed in the literature for content classification usually learn models considering textual and contextual features from real and fake news to minimize the spread of disinformation. One of the prominent approaches to detect fake news is One-Class Learning (OCL), as it minimizes the data labeling effort, requiring only the labeling of fake news documents. The performance of these algorithms depends on the structured representation of the documents used in the learning process. Generally, a textual-based unimodal representation is used, such as bag-of-words or representations based on linguistic categories. We propose MVAE-FakeNews, a multimodal representation method to detect fake news in OCL. The proposed approach uses a Multimodal Variational Autoencoder, learns a new representation from the combination of two modalities considered promising for fake news detection: text embeddings and topic information. In the experiments, we used three datasets considering Portuguese and English languages. Results show that the MVAE-FakeNews obtained a better F1-Score for the class of interest, outperforming another nine methods in ten of twelve evaluated scenarios. MVAE-FakeNews presented a better average ranking and statistical difference from other representation models. The proposed method proved to be promising to represent the texts in the OCL scenario to detect fake news.

Learning Textual Representations from Multiple Modalities to Detect Fake News Through One-Class Learning

The dissemination of news through digital media has amplified Fake News proliferation. In the face of this scenario, sentiment-based methods have presented promising results in Fake News detection. Although sentiment-based methods can extract sentiment (i.e., po- larity and/or emotion) from either texts or images available in news, the ones applied to Portuguese-written news have considered sen- timent exclusively extracted from texts. Thus, this study proposes a multimodal method that, besides the polarity and emotions ex- tracted from texts, also considers sentiment extracted from news’ images in order to detect Fake News written in Portuguese. The proposed method showed promising results in experimental data, overcoming the baseline methods in 8 p.p.

Downloads

Next from WebMedia 2021

Systematic mapping of the literature on mobile apps for people with Autistic Spectrum Disorder

Similar lecture

TA2N: Two-Stage Action Alignment Network for Few-Shot Action Recognition

.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

Systematic mapping of the literature on mobile apps for people with Autistic Spectrum Disorder

Similar lecture

TA2N: Two-Stage Action Alignment Network for Few-Shot Action Recognition

Downloads