Papers

Basic information

Name ENDO KAZUKI
Belonging department Management Information Course,Department of Business,Faculty of Humanities and Social Sciences
Occupation name Associate Professor
researchmap researcher code
researchmap agency

Title 

Degraded image classification using knowledge distillation and robust data augmentations (accepted, to appear)

					

					

						

							

								

									
Bibliography怀Type 

									
								

							

						

						
Joint Author

					

					

						

							

								

									
Summary 

									
								

							

						

						
This paper proposes an effective combination of data augmentations to train the classification network of degraded images using knowledge distillation.
The results show that our proposed method outperforms existing methods regarding the interval mean accuracy of all degradation levels.
Dinesh Daultani, Masayuki Tanaka, Masatoshi Okutomi, Kazuki Endo

					

					

						

							

								

									
Magazine(name) 

									
								

							

						

						
IEICE Transactions on Information and Systems

					

					

						

							

								

									
Date of Issue 

									
								

							

						

						
202412