6 papers · 1 filter
Improved Encoding for Overfitted Video Codecs
Thomas Leguay, Théo Ladune, Pierrick Philippe +1
Overfitted neural video codecs offer a decoding complexity orders of magnitude smaller than their autoencoder counterparts. Yet, this low complexity comes at the cost of limited co…
Upsampling Improvement for Overfitted Neural Coding
Pierrick Philippe, Théo Ladune, Gordon Clare +3
Neural image compression, based on auto-encoders and overfitted representations, relies on a latent representation of the coded signal. This representation needs to be compact and…
Cool-chic video: Learned video coding with 800 parameters
Thomas Leguay, Théo Ladune, Pierrick Philippe +1
We propose a lightweight learned video codec with 900 multiplications per decoded pixel and 800 parameters overall. To the best of our knowledge, this is one of the neural video co…
Cool-Chic: Perceptually Tuned Low Complexity Overfitted Image Coder
Théo Ladune, Pierrick Philippe, Gordon Clare +2
This paper summarises the design of the Cool-Chic candidate for the Challenge on Learned Image Compression. This candidate attempts to demonstrate that neural coding methods can le…
ED: Perceptually tuned Enhanced Compression Model
Pierrick Philippe, Théo Ladune, Stéphane Davenet +1
This paper summarises the design of the candidate ED for the Challenge on Learned Image Compression 2024. This candidate aims at providing an anchor based on conventional coding te…
Low-complexity Overfitted Neural Image Codec
Thomas Leguay, Théo Ladune, Pierrick Philippe +2
We propose a neural image codec at reduced complexity which overfits the decoder parameters to each input image. While autoencoders perform up to a million multiplications per deco…