Wavelets in the Deep Learning ERA - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2021

Wavelets in the Deep Learning ERA

Abstract

Sparsity based methods, such as wavelets, have been state-of-the-art for more than 20 years for inverse problems before being overtaken by neural networks. In particular, U-nets have proven to be extremely effective. Their main ingredients are a highly non-linear processing, a massive learning made possible by the flourishing of optimization algorithms with the power of computers (GPU) and the use of large available data sets for training. While the many stages of non-linearity are intrinsic to deep learning, the usage of learning with training data could also be exploited by sparsity based approaches. The aim of our study is to push the limits of sparsity with learning, and comparing the results with U-nets. We present a new network architecture, which conserves the properties of sparsity based methods such as exact reconstruction and good generalization properties, while fostering the power of neural networks for learning and fast calculation. We evaluate the model on image denoising tasks and show it is competitive with learning-based models.
Fichier principal
Vignette du fichier
learnlets_eusipco_2020 (6).pdf (3.3 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03020214 , version 1 (23-11-2020)

Identifiers

  • HAL Id : hal-03020214 , version 1

Cite

Zaccharie Ramzi, Jean-Luc Starck, Thomas Moreau, Philippe Ciuciu. Wavelets in the Deep Learning ERA. EUSIPCO 2020 - 28th European Signal Processing Conference, Jan 2021, Amsterdam, Netherlands. ⟨hal-03020214⟩
268 View
673 Download

Share

Gmail Facebook X LinkedIn More