Creating and sharing knowledge for telecommunications

Project: End-to-End Deep learning based image Compression

Acronym: DRAGON
Main Objective:
The DRAGON (Deep leaRning bAsed imaGe cOmpressioN) project’s aim is to develop state-of-the-art deep learning methods that are able to determine a compact image representation model, obtained (learned) from a large amount of visual data, and are capable of representing the wide variety of visual content that is available today with a high value of compression ratio. This project will advance existing solutions in the learning-based image coding field, which already show encouraging results in terms of rate- distortion performance, especially in comparison with conventional image codecs (e.g. JPEG 2000 and HEVC Intra) that compress the visual information with hand-crafted transforms, entropy coding and quantization schemes.
Reference: 0001
Funding: IT
Start Date: 01-10-2021
End Date: 01-10-2023
Team: João Miguel Duarte Ascenso, Catarina Isabel Carvalheiro Brites Ascenso, Nuno Miguel Morais Rodrigues, Luís Filipe Barbosa de Almeida Alexandre
Groups: Multimedia Signal Processing – Lx, Multimedia Signal Processing – Lr, Pattern and Image Analysis – Cv
Partners: IT
Local Coordinator: João Miguel Duarte Ascenso

Associated Publications