SG2002 is a high-performance, low-power chip designed for various product fields such as edge intelligent surveillance IP cameras, smart door locks, visual doorbells, and home intelligence. It integrates H.264 video compression and decoding, H.265 video compression encoding, and ISP capabilities. It supports multiple image enhancement and correction algorithms such as HDR wide dynamic range, 3D noise reduction, defogging, and lens distortion correction, providing customers with professional-grade video image quality.

The chip also incorporates a self-developed TPU, delivering 1.0 TOPS of computing power under 8-bit integer operations. The specially designed TPU scheduling engine efficiently provides high-bandwidth data flow for all tensor processing unit cores. Additionally, it offers users a powerful deep learning model compiler and software SDK development kit. Leading deep learning frameworks like Caffe and Tensorflow can be easily ported to its platform. Furthermore, it includes security boot, secure updates, and encryption, providing a series of security solutions from development, mass production, to product applications.

The chip integrates an 8-bit MCU subsystem, replacing the typical external MCU to achieve cost-saving and power efficiency goals.