Video Technology Research Group (ViTech)

The WebMedia conference, organized annually by the Brazilian Computer Society (SBC), is the premier event on multimedia and web systems in Brazil, offering an excellent opportunity for scientific and technical exchange among students, researchers, and professionals in the fields of Multimedia, Hypermedia, and the Web. This year, the event will be even more special as we celebrate the 30th edition of WebMedia, which will take place from October 14th to 18th at the Federal University of Juiz de Fora (UFJF) in Minas Gerais. 📄 Approved papers: Multi-Domain Spatio-Temporal Deformable Fusion Model for Video Quality Enhancement Authors: Garibaldi da Silveira Júnior, Gilberto Kreisler, Bruno Zatt, Daniel Palomino, Guilherme Corrêa Fast ISP Mode Decision for the VVC Intra Prediction Using Machine Learning Authors: Larissa Araújo, Adson Duarte, Guilherme Corrêa, Bruno Zatt, Daniel Palomino A Machine-Learning-Driven Fast Video-Based Point Cloud Compression (V-PCC) Authors: Gustavo Rehbein, Eduardo Costa, Guilherme Corrêa, Cristiano dos Santos, Marcelo Porto A Coding-Efficiency-Aware Fast Heuristic for VVC Intra-Prediction Targeting 360° Videos Authors: Rogério Cassanta Rosado, Otávio Santos, Fernandes dos Santos, Franklin Sales de Oliveira, Lucas Superti da Silva, Gustavo Sanchez, Daniel Palomino, Luciano Volcan Agostini Predição Intra-Quadro Baseada em Aprendizado Profundo para Light Fields Densos Authors: Ítalo Machado, Daniel Palomino, Attilio Fiandrotti (UniTo), Bruno Zatt This is a significant milestone for our team, and we are excited to present our research and exchange knowledge with other experts in the field. See you in Juiz de Fora!

🌟 Approved Projects - Rio Grande do Sul Research Support Foundation (FAPERGS)🌟 We are pleased to announce the approval of projects under the Funding Call 09/23 (PqG). The approval of these projects is an important achievement for ViTech, enabling the acquisition of new equipment and strengthening our research capabilities. 🔍 Approved Projects: Green Codec: Energy-Efficient Video Coding and Quality Enhancement with Machine Learning Coordinator: Prof. Guilherme Corrêa Dedicated Hardware Design for Visual Signal Compression Using Machine Learning Coordinator: Prof. Marcelo Porto 🏢 The approved projects represent funding for the acquisition of equipment for the ViTech laboratory, enabling the publication of articles and the participation of our researchers in scientific events. Both projects have a duration of 3 years. #ViTech #FAPERGS #PesquisadorGaúcho #Innovation #Research #Science #Technology

🎥🔧 Why Do We Need Dedicated Hardware for Video Encoding? Modern smartphones are an excellent example of how dedicated hardware is essential. They have dedicated circuits specifically designed to efficiently process the images captured by cameras with high quality. 📱 Understand the importance of dedicated hardware and how it enhances the user experience on mobile devices! 📲 #ViTech #DecodingViTech #DedicatedHardware #VideoCompression #Technology #Innovation

🎥🔧 Interframe Prediction in Video Encoding: What is it and how does it work? This technique aims to exploit the similarities between neighboring frames in a video, in order to reduce the required storage space and the bandwidth for transmission. Learn what it is, how it works, and what the challenges are for this essential task in the smooth functioning of a video codec! 📉🔍 #ViTech #DecodingViTech #InterframePrediction #VideoCompression #Technology #Innovation

🔬✨ REDUCING COMPLEXITY IN VIDEO TRANSCODERS ✨🔬 Today, we are pleased to share a summary of an interview with two of our exceptional researchers, Alex Borges and Caroline Camargo. They are leading the research on video transcoding here at ViTech. But what is transcoding? It is the process of converting video files, adjusting characteristics such as resolution and image quality, or switching between different encoding formats (see more details in our previous post). The main goal of Alex and Caroline is to optimize this process, making it faster and more energy-efficient without sacrificing visual quality or user experience. ⚙️ To learn more, check out Alex Borges' Ph.D. thesis, which is available at: https://lnkd.in/gXvukfAs #ViTech #VideoCoding #Innovation #Technology #Transcoding #Research

🔍 Decoding ViTech 🔍 Today we're going to talk about an essential process in the world of streaming: Transcoding! 👩🏼💻 Have you ever wondered how YouTube or Netflix manage to deliver high-quality videos on various devices and internet connections? The secret lies in transcoding! This fundamental service adapts videos to different resolutions and formats, optimizing the user experience and allowing for uninterrupted streaming. In the world of videos, transcoding is the key to efficient and high-quality broadcasting! 🔑🎥 Stay tuned for more terms decoded here in our "Decoding ViTech"! #ViTech #DecodingViTech #Transcoding #Streaming #Technology #Innovation

We are proud to share our accepted works for the 37th Symposium on Integrated Circuits and Systems Design (SBCCI 2024), which will take place during Chip on the Cliffs, in João Pessoa, Paraíba (Brazil). 1) Bit-Width Optimized Transposition Buffer Design for the AV1 2D-DCT Transform Authors: Jelson Rodrigues, Jones Goebel, Bruno Zatt, Luciano Agostini, Marcelo Porto 2) A Real-Time UHD 4K Hardware for VVC Affine Linear Equation System Solving Authors: Denis Maass, Marcello Muñoz, Murilo Perleberg, Luciano Agostini, Marcelo Porto 3) A Hardware-Friendly Acceleration of VVC Affine Motion Estimation Using Decision Trees Authors: Ramiro Viana, Marta Loose, Rafael Ferreira, Marcelo Porto, Guilherme Corrêa, Luciano Volcan Agostini Chip on the Cliffs will bring together researchers from around the world in the fields of EDA, circuit and system design, and testing. In addition to technical sessions, the event will feature tutorials, panels, an exhibition, and working group meetings. This year's edition will be held in João Pessoa, Paraíba (Brazil), from September 2 to 6, 2024.

This week our group attended the 15th IEEE LASCAS, in Punta del Este (Uruguay). LASCAS is the international symposium and flagship event of the IEEE Circuits and Systems Society in Latin America. This was an exceptional week with a lot of learning, exchange and networking. ViTech was present with a group composed of 20 members (students and professors), who presented 10 original research papers. The paper "Speeding Up the AV1 Global Warped Motion Compensation", authored by William Kolodziejski, Robson Domanski, and Luciano Volcan Agostini, received the Honorable Mention Award.

ViTech members have achieved notable results with the acceptance of 10 papers authored or co-authored by them at the upcoming 2024 IEEE Latin America Symposium on Circuits and Systems (LASCAS). LASCAS stands as the flagship international symposium organized by the IEEE Circuits and Systems Society in Latin America. LASCAS is scheduled to take place in Punta del Este, Uruguay, from February 27 to March 1, 2024. Below, you will find the titles and authors of the accepted papers. 1 - A Hardware-Friendly Fast VVC Test Zone Search Algorithm Using Machine Learning, by Ramiro Viana, @Fernando Sagrilo, Rafael Ferreira, @Marta Loose, Marcelo Porto, Guilherme Corrêa, and Luciano Agostini 2 - Multi-Codec Video Quality Enhancement Model Based on Spatio-Temporal Deformable Fusion, by Gilberto Kreisler, Garibaldi da Silveira Junior, Bruno Zatt, Daniel Palomino and Guilherme Corrêa 3 - Coding Efficiency and Time Evaluation of Apple A15 Bionic Chipset HEVC Encoder, by Vítor Costa, Murilo Perleberg, Luciano Agostini, Marcelo Porto 4 - Coding Efficiency and Complexity Analysis of the Geometry-based Point Cloud Encoder, by Cristiano Santos, Leandro Tavares, Eduardo Costa, Gustavo Rehbein, Guilherme Corrêa and Marcelo Porto 5 - Error-Resilience Profiling of Inter-Frame Prediction at VVC Encoders for Approximate Storage, by Yasmin Souza Camargo, Matheus Isquierdo, Renira Soares, Bruno Zatt, Daniel Palomino and Felipe Sampaio 6 - Speeding Up the AV1 Global Warped Motion Compensation, by William Kolodziejski, Robson Domanski and Luciano Agostini 7 - Alternative Reference Samples to Improve Coding Efficiency for Parallel Intra Prediction Solutions, by Iago Storch, Nuno Roma, Daniel Palomino, Sergio Bampi 8 - Neural Network Based Light Field Predictors: An Evaluation, by Matheus Jahnke, Italo Machado, Lucas Ikenoue and Bruno Zatt 9 - Rate-Distortion and Complexity Analysis of Fast Video Encoders, by Eduardo Ferraz Lodi, Arthur Scarpatto, Ismael Seidel, Guilherme Correa and Mateus Grellert 10 - Computational Effort Reduction of VVC Inter-Frame Prediction: A Literature Review, by Marta Breunig Loose, Ramiro Viana, Guilherme Corrêa and Luciano Agostini