Iaroslav Bespalov
Research Subjects
Iaroslav Bespalov has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Dmitry V. Dylov
Artificial Intelligence Research Institute, Moscow, Russian Federation and Skolkovo Institute of Science and Technology, Moscow, Russian Federation
3 shared publications Ivan Zakazov
1 shared publication- Nazar Buzun1 shared publication
- Viktor Shipitsin
Center for Artificial Intelligence Technologies, Skolkovo Institute of Science and Technology, Moscow, Russia
1 shared publication Vladimir Shaposhnikov
1 shared publication
Latest Publications
- 1 August 2022
- Published by Elsevier in Computer Vision and Image Understanding
- Vol. 223
GAFL: Global adaptive filtering layer for computer vision
- 1 August 2022
- Published by Elsevier in Computer Vision and Image Understanding
- Vol. 223
GAFL: Global adaptive filtering layer for computer vision
- Open Access
Feather-Light Fourier Domain Adaptation in Magnetic Resonance Imaging
Generalizability of deep learning models may be severely affected by the difference in the distributions of the train (source domain) and the test (target domain) sets, e.g., when the sets are produced by different hardware. As a consequence of this domain shift, a certain model might perform well on data from one clinic, and then fail when deployed in another. We propose a very light and transparent approach to perform test-time domain adaptation. The idea is to substitute the target low-frequency Fourier space components that are deemed to reflect the style of an image. To maximize the performance, we implement the "optimal style donor" selection technique, and use a number of source data points for altering a single target scan appearance (Multi-Source Transferring). We study the effect of severity of domain shift on the performance of the method, and show that our training-free approach reaches the state-of-the-art level of complicated deep domain adaptation models. The code for our experiments is released.
Open AccessFeather-Light Fourier Domain Adaptation in Magnetic Resonance Imaging
Generalizability of deep learning models may be severely affected by the difference in the distributions of the train (source domain) and the test (target domain) sets, e.g., when the sets are produced by different hardware. As a consequence of this domain shift, a certain model might perform well on data from one clinic, and then fail when deployed in another. We propose a very light and transparent approach to perform test-time domain adaptation. The idea is to substitute the target low-frequency Fourier space components that are deemed to reflect the style of an image. To maximize the performance, we implement the "optimal style donor" selection technique, and use a number of source data points for altering a single target scan appearance (Multi-Source Transferring). We study the effect of severity of domain shift on the performance of the method, and show that our training-free approach reaches the state-of-the-art level of complicated deep domain adaptation models. The code for our experiments is released.
- 27 May 2022
- Published by Elsevier in Pattern Recognition
- Vol. 131
BRULÈ: Barycenter-Regularized Unsupervised Landmark Extraction
- 27 May 2022
- Published by Elsevier in Pattern Recognition
- Vol. 131
BRULÈ: Barycenter-Regularized Unsupervised Landmark Extraction