We are engaged in the study of computer vision and its related fields, including image processing, machine learning, and natural language processing. Computer vision aims to develop artificial intelligence capable of perceiving, identifying, and making decisions about various visual phenomena, ranging from image sensing to semantic recognition. To achieve this objective, we investigate theoretical and practical issues in computer vision, such as material recognition, urban scene modeling, deep neural networks, probabilistic graphical models, artificial neural networks for neuroscience, visual fashion analytics, and attribute perception in natural language. Currently, we are concentrating on the theoretical analysis and practical application of deep neural networks due to the recent advancements in this field. For instance, we investigate how deep learning models can effectively identify and recognize material properties beyond object categories (see Fig. 1), and how artificial neural networks relate to human perception. Furthermore, we have researched methods for comprehending visual information in images using natural language (see Fig. 2).