This paper introduces a novel framework, Gesture and Appearance Cutout Embedding (GACE), that supports real-time integration of human appearance and gesture-guided control within a game. It aims to enhance immersion since it allows game users to see their personal appearance in a real-time manner. In addition, we exploit the gesture-based control to allow user interaction with other in-game characters. With the goal to make implementation easier, we address the challenges in the whole pipeline of video processing, gesture recognition, and communication. The system is successfully integrated into both entertainment and serious games. Extensive experiments show that the proposed system runs reliably and comfortably with a commodity setting. Meanwhile, the user impression study indicates our system is favored by end users.