An Image Emotion Classification Method Based on Supervised Contrastive Learning.

This article proposes an image emotion classification model based on supervised contrastive learning, which includes two modules of low-level feature extraction and deep emotion feature extraction, and feature fusion is used to enhance the overall perception of image emotion. In the low-level feature extraction module, the LBP (Local Binary Patterns) algorithm is used to extract texture features of images. In the deep emotion feature extraction module, supervised contrastive learning is introduced to narrow the intra-class distance between the same emotion images and expand the inter-class distance between different emotion images, so as to improve the classification performance of the overall model. Comprehensive experiments on the FI emotion dataset show that the proposed model has good performance for image emotion classification.