Comparison of Image-Based Cultural Heritage Recognition Techniques: Few-Shot Learning vs. Image Retrieval

Museums and art galleries try many different ways to get visitors interested in the exhibits shown. One widely used method is to recognize images of exhibits taken with mobile devices and provide relevant information to visitors. However, in order for this method to be effective, it is necessary to be able to recognize the objects in the photographed images with a high degree of accuracy. In this paper, we compare the recognition performance of captured images between few-shot learning and image retrieval, especially when cultural heritages with small quantities of various types are the main exhibits, and determine which technique is more effective in the cultural heritage field. By analyzing the experimental results in this paper, we can identify the strengths and weaknesses of each technique, and these results can be used to determine which technique should be used in actual exhibition situations.