Alan Livio Vasconcelos Guedes

Alan Livio Vasconcelos Guedes

Lecturer, University of Reading

About me

I am a Lecturer at the University of Reading. I am working on leveraging machine learning for optimizing multimedia systems pipelines for both traditional and immersive (360 videos, point clouds, meshs, etc.) media technologies. Previously, I was a postdoc at UCL working on immersive media and a postdoc at PUC-Rio working on video analysis and interactive TV, I acted team-leading or PI in funded research projects, contributed to ACM SIGMM communities and tandard committees (Brazilian TV Forum and ITU-T SG16). I also seek to get involved in purpose-driven projects with social impact.

Interests
  • Multimedia Systems
  • Machine Learning

Experience

 
 
 
 
 
University of Reading
Lecturer Researcher
March 2024 – Present Reading, UK
  • Contribute to Departament Teaching and Research
  • Supervised undergrad and postgraduate students.
 
 
 
 
 
UCL
Postdoc Researcher
April 2021 – March 2024 Reading, UK
  • Contribute to ACM SIGMM committees and workshop organiser;
  • Co-supervised undergrad and M.Sc. students.
  • SpheryStream project funded by EPSRC to Improve 360-video streaming with profile-based methods
 
 
 
 
 
PUC-Rio
Postdoc Researcher
October 2016 – January 2021 Rio de Janeiro, Brazil
  • Co-supervised undergrad and M.Sc. students.
  • VideoMR project, funded by Microsoft & RNP to detect inappropriate content in video scenes;
  • AML (Anti-Money Laundry) project, funded by BTG Bank, to detect financial transaction risk and anomaly detection.
  • GT-RAP project, funded by RNP, to develop a blockchain-based system for academic certificates.
  • SBTVD Conformance Test Suite for Ginga D profile, funded by the Brazilian Ministry of Science and Technology, to be used by TV manufacturers
 
 
 
 
 
PUC-Rio, LAVID Lab
PhD Researcher
March 2013 – September 2017 Rio de Janeiro, Brazil
  • Thesis used in international standards.
  • PUC-Rio Ginga implementation, funded by PUC-Rio Software Licenses, to be used by TV manufacturers;
  • SBTVD Conformance Test Suite for Ginga A/B/C profiles, funded by the Brazilian Ministry of Science and Technology, to be used by TV manufacturers.
 
 
 
 
 
Dynavideo
Research Engineer
October 2012 – January 2013 João Pessoa, Brazil
  • Brasil4D project, funded by the World Bank, to transmit interactive TV content with social features to low-income families.
 
 
 
 
 
UFPB, LAVID Lab
M.Sc. Researcher
October 2010 – September 2012 João Pessoa, Brazil
  • GingaStore, funded by RNP, to design a broadband system to deliver TV applications.
  • Art TVDI, funded by RNP, to design agile methods for TV application development.

Research Interests

Users’ consumption of multimedia content grows exponentially in the Internet, smart devices, social media, and other contexts. Multimedia Systems addresses technological issues for enabling and improving consumption, such as content delivery, adaptation, accessibility and personalisation, and others. See some aspects of my multimedia research below.

Multimedia Systems

Multimedia systems deal with the delivery of multimedia data in digital networks. A multimedia system infrastructure requires components such as protocols, services, middleware, and mechanisms to handle a well-defined quality of service (QoS).

Some of my contributions in the field are in TV middleware:

For me, some important researchers in this field are: 🎓 Klara Nahrstedt 🎓 Christian Timmerer

Immersive Media

Immersive media research aims to make multimedia systems more natural and with higher QoE, especially by imitating the physical world through digital simulation. I particularly researched it through VR/360 video and multisensorial aspects.

Some of my contributions in the field are in 360 videos:

For me, some important researchers in this field are: 🎓 Pablo Cesar, CWI, 🎓 Jesús Gutiérrez 🎓 Cornelius Hellge

Video Understanding

Video Understanding deals with how computers can gain high-level understanding from digital images or videos. Unlike Multimedia Systems’ aspects of delivery and feasibility, it aims to analyse multimedia data automatically. Usage scenarios are video retrieval and browsing, anomaly detection, and others.

Some of my contributions in the field are unsupervised video analysis:

For me, some important researchers in this field are: 🎓 Richard Szeliski 🎓 Shaogang Gong

Social Impact

I truly believe technology can be a vehicle for social impact, especially in multimedia. Today, multimedia/web/video data are spread and are useful for accessibility, governance transparency, minority inclusion, social justice/activism, etc.

Recent Publications

Quickly discover relevant content by filtering publications.
(2021). A Cluster-Based Method for Action Segmentation Using Spatio-Temporal and Positional Encoded Embeddings. WebMedia ‘21: Brazilian Symposium on Multimedia and the Web, Belo Horizonte, Minas Gerais, Brazil, November 5-12, 2021.

Cite DOI URL

(2021). Quality Enhancement of Highly Degraded Music Using Deep Learning-Based Prediction Models for Lost Frequencies. WebMedia ‘21: Brazilian Symposium on Multimedia and the Web, Belo Horizonte, Minas Gerais, Brazil, November 5-12, 2021.

Cite DOI URL

Contact

  • Pepper Lane, Whiteknights, Reading, RG6 6DH
  • Room 142, Polly Vacher