- Style transfert using semi-dual formulation for minimizing the OT distance between VGG-19 features:
- Textures barycenters using our texture generation algorithm (Alg.1 from this work).
The key idea is to generate a new texture such that its patch distributions at various scales are Wasserstein barycenters of the patch distribution of the two inputs:
joint work with Arthur Leclaire, Nicolas Papadakis and Julien Rabin
Abstract: In this paper, we propose a framework to train a generative model for texture imagesynthesis from a single example. To do so, we exploit the local representationof images via the space of patches, that is, square sub-images of fixed size (e.g. 4×4). Our main contribution is to consider optimal transport to enforce themultiscale patch distribution of generated images, which leads to two differentformulations. First, a pixel-based optimization method is proposed, relying ondiscrete optimal transport. We show that it is related to a well-known textureoptimization framework based on iterated patch nearest-neighbor projections, whileavoiding some of its shortcomings. Second, in a semi-discrete setting, we exploitthe differential properties of Wasserstein distances to learn a fully convolutionalnetwork for texture generation. Once estimated, this network produces realisticand arbitrarily large texture samples in real time. The two formulations result innon-convex concave problems that can be optimized efficiently with convergenceproperties and improved stability compared to adversarial approaches, withoutrelying on any regularization. By directly dealing with the patch distribution ofsynthesized images, we also overcome limitations of state-of-the art techniques,such as patch aggregation issues that usually lead to low frequency artifacts (e.g. blurring) in traditional patch-based approaches, or statistical inconsistencies (e.g. color or patterns) in learning approaches.
I’m presenting the work
at the European Signal Processing Conference EUSIPCO 2019
The slides can be found herafter: presentation_EUSIPCO
I’m presenting the work
in the session O3.1 Restauration, déconvolution, débruitage at the GRETSI 2019 Colloque
The slides can be found herafter: presentation Gretsi 2019
I presented a poster [poster_smai19] during the 9e biennale de la SMAI
J’ai présenté le cours
le jeudi 18 avril 2019, dans le cadre du module image du LIRMM à Montpellier.
I’m giving a talk for the Cambridge Image Analysis Seminars, on
Thursday, March 21st, 11:00
Je participerai à la Cérémonie des Prix de la Fondation Mines-Télécom 2019 le mercredi 27 mars. Pour cette occasion, j’ai fait une vidéo résumant ma thèse en 3 minutes :
J’ai reçu le 3e prix de thèse lors de cette cérémonie :
Je tiens à remercier la fondation Mines-Télécom ainsi que le programme de financement de thèses Futur & Ruptures !
I’m giving a talk for the Young Researchers in Imaging Seminars, on
Wednesday, February 27th, 15:00
in Amphi Darboux at IHP.
Title: On the use of Gaussian models on patches for image denoising
Abstract: Some recent denoising methods are based on a statistical modeling of the image patches. In the literature, Gaussian models or Gaussian mixture models are the most widely used priors.
In this presentation, after introducing the statistical framework of patch-based image denoising, I will propose some clues to answer the following questions: Why are these Gaussian priors so widely used? What information do they encode?
In the second part, I will present a mixture model for noisy patches adapted to the high dimension of the patch space. This results in a denoising algorithm only based on statistical tools, which achieves state-of-the-art performance.
Finally, I will discuss the limitations and some developments of the proposed method.