On approaches to solving the problem of attribution of works by representatives of the Russian avant-garde through the implementation of image recognition methods
Julia A. Brodskaya1; 1Gagarin's State Technical University of Saratov, Saratov, Russia
Abstract
The paper discusses the possibility of practical application of image recognition methods in attributing paintings of selected representatives of the Russian avant-garde (using the example of Natalia Goncharova's painting and graphics, taking into account the peculiarities of her manner of execution and handwriting, in general, in different periods of creativity). This application of the theory shows one of the most relevant areas of application of deep learning and computer vision. The PyTorch machine learning framework was used in the study. Due to the fact that Natalia Goncharova's works began to be forged especially often after the sale of her painting in 2010 at Christie's auction and the prices of her works increased significantly, the relevance of the study is particularly significant.
Speaker
Julia Anatolyevna Brodskaya
Gagarin's State Technical University of Saratov
Russia
Discussion
Ask question
