Hand Pose Estimation from RGB Images Based on Deep Learning: A Survey

With the development of computing technique, the emergence of computers and their derivatives as the carrier of modern artificial intelligence has penetrated into people's daily life from all aspects, and the status of human-computer interaction has become increasingly prominent. As hand is the main operating tool of human beings, its position and orientation in space are crucial for many potential applications, such as the occasion of interacting with VR devices. An important process of hand gesture recognition is hand pose estimation. Still, there are many difficulties in 3D hand pose estimation from a single RGB image. The rise of big data, the emergence of neural networks and the iteration of high computing power equipment have led to the emergence of deep learning in vision fields, and the application of hand pose estimation has made great breakthroughs. This paper briefly introduces the methods of hand pose estimation from RGB images based on deep learning, summarizes the existing research results and makes an outlook on the future development trend of the field.