Applied machine learning for image coding

Objectives: - To explore some of the recent methods of machine learning for image coding (compression) - To design a combination of machine learning method for Image coding compression. Description: Video traffic constituted will constitute 82% of all IP traffic by 2022. Improving image and video coding methods and video networking schemes is therefore vital to cope with this increasing demand. In recent years, we have witnessed how Artificial Intelligence (AI) and Machine Learning (ML) revolutionized the field of image and video coding and streaming as these new solutions now offer state-ot-the-art in many high-level and low-level image and video related tasks. Notes: - A combinations of Convolutional Neural Networks (CNN)-based image coding tools ( intra coding, rate-distortion optimization, deblocking filters, interpolation filters, chroma from luma prediction methods) need to be used to design a machine learning image compression. For more techniques, check these two links: - [login to view URL] - [login to view URL] ( propose your own combination ) - The combination must give a high-quality image after compressing. - The design need to be done using Google Colab. - MS Word file that explains the code in details is required.