- Afyon Kocatepe Üniversitesi Fen Ve Mühendislik Bilimleri Dergisi
- Vol: 23 Issue: 3
- End-to-End Artworks Generation Via Deep Convolutional Based Generative Adversarial Networks
End-to-End Artworks Generation Via Deep Convolutional Based Generative Adversarial Networks
Authors : Nazlı Turhan, Ahmet Haşim Yurttakal
Pages : 671-676
Doi:10.35414/akufemubid.1269356
View : 30 | Download : 26
Publication Date : 2023-06-28
Article Type : Research Article
Abstract :While artificial intelligence (AI) technologies are used in many fields such as health, education, art and continue to develop rapidly, emerging artificial intelligence solutions are also being addressed by different disciplines, such as informatics and law. Apart from the problems of legal rules\' having access to the speed of social change, the search of a legal infrastructure that is suitable for keeping up with these changes has started to make itself felt in recent years. In the study, the technical stages of digital artworks created by using contentious producer networks from deep learning algorithms were discussed and evaluated within the scope of intellectual and artistic works law. In the study, 6989 abstract and portrait paintings, which are a subset of the Wiki-Art dataset, were used. As a result, it has been seen that the number of images in the dataset affects the originality of the outputs. It is thought that the proposed method can be applied to different branches of art and can give art lovers a different perspective.Keywords : Sanat Eserleri, Çekişmeli Üretici Ağlar, Uçtan uca, Yapay Zeka