Arthur Floquet
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Arthur Floquet has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Duong-Hung Pham
IRIT Lab, Université de Toulouse, and CNRS UMR 5505, 118 Route de Narbonne, Toulouse, 31400, Occitanie, France
2 shared publications - Emmanuel Soubies
IRIT UMR CNRS Toulouse INP Université de Toulouse 5505 31000 Toulouse France
2 shared publications Denis Kouame
IRIT Laboratory, Universit de Toulouse, Toulouse, France
1 shared publication- Denis Kouame
Université de Toulouse, CNRS, IRIT 5 , Toulouse 31042
1 shared publication - Sayantan Dutta
Department of Radiology, Weill Cornell Medicine, New York, NY, USA
1 shared publication
Latest Publications
- 30 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
- Vol. 72 (9) , 1235-1244
Spatially Variant Ultrasound Image Restoration With Product Convolution
The process of ultrasound (US) image formation can generally be modeled, using a linear and shift invariance approximation, as a convolution. In practice, the point spread function (PSF) is shift-variant. Here, we consider the restoration problem using a shift-variant PSF, where it is modeled as a product convolution. We argue that the US PSF varies smoothly enough for product convolution to serve as an efficient and effective direct model for US image restoration. We present a strategy for constructing the product-convolution operator and derive an efficient optimization scheme. We finally validate our approach on both simulated and real data, demonstrating state-of-the-art results while achieving significantly faster processing times.
- 30 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
- Vol. 72 (9) , 1235-1244
Spatially Variant Ultrasound Image Restoration With Product Convolution
The process of ultrasound (US) image formation can generally be modeled, using a linear and shift invariance approximation, as a convolution. In practice, the point spread function (PSF) is shift-variant. Here, we consider the restoration problem using a shift-variant PSF, where it is modeled as a product convolution. We argue that the US PSF varies smoothly enough for product convolution to serve as an efficient and effective direct model for US image restoration. We present a strategy for constructing the product-convolution operator and derive an efficient optimization scheme. We finally validate our approach on both simulated and real data, demonstrating state-of-the-art results while achieving significantly faster processing times.
- 16 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 31 (10709908) , 381-385
Automatic Tuning of Denoising Algorithms Parameters Without Ground Truth
Denoising is omnipresent in image processing. It is usually addressed with algorithms relying on a set of hyperparameters that control the quality of the recovered image. Manual tuning of those parameters can be a daunting task, which calls for the development of automatic tuning methods. Given a denoising algorithm, the best set of parameters is the one that minimizes the error between denoised and ground-truth images. Clearly, this ideal approach is unrealistic, as the ground-truth images are unknown in practice. In this work, we propose unsupervised cost functions—i.e., that only require the noisy image—that allow us to reach this ideal gold standard performance. Specifically, the proposed approach makes it possible to obtain an average PSNR output within less than 1% of the best achievable PSNR.
- 16 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 31 (10709908) , 381-385
Automatic Tuning of Denoising Algorithms Parameters Without Ground Truth
Denoising is omnipresent in image processing. It is usually addressed with algorithms relying on a set of hyperparameters that control the quality of the recovered image. Manual tuning of those parameters can be a daunting task, which calls for the development of automatic tuning methods. Given a denoising algorithm, the best set of parameters is the one that minimizes the error between denoised and ground-truth images. Clearly, this ideal approach is unrealistic, as the ground-truth images are unknown in practice. In this work, we propose unsupervised cost functions—i.e., that only require the noisy image—that allow us to reach this ideal gold standard performance. Specifically, the proposed approach makes it possible to obtain an average PSNR output within less than 1% of the best achievable PSNR.