Author recognition of Fine-Art paintings

With the continuous development of digitalization, the automated processing and analysis of paintings is becoming increasingly important. The automatic author recognition of Fine-Art paintings will play an important role in the further development of the painting digitization. This paper proposes a densely connected convolutional network structure with multi-layer feature fusion to automatically identify the author of the painting. Unlike traditional CNN, Multi-layer Feature Fusion DenseNet (MFDN) combines shallow features with deep features to make final decisions, making recognition results more accurate. The article proves through experiments that our models is better than the existing state-of-the-art models in this field.