The institute aims to build an end-to-end smarting imaging system with multiple sensors and generative processing techniques. It combines the power of new hardware sensors, software processing ability using generative powers, and a complete evaluation system, and can adapt to different tasks. Specifically, it includes the following 3 components: 1) A camera system, including different number of sensors, different types of sensors, and different sensor configuration. Therefore, it can easily adapt to different use cases. 2) A generative processing system that can fuse the information from a single or multiple sensors, and handle difficult scenario, like lowlight or noise. 3) An evaluation pipeline that aligns with human preferences.