Core Technology
Based on the foundation of proprietary technologies and a core “brain” built on a deep learning platform,
SenseTime has rapidly opened up AI application in multiple vertical scenarios.

1
Visual Perception
With SenseTime’s deep learning and computer vision technology, the core algorithms such as high-performance lightweight target detection, tracking and recognition, was widely adopted in the Driver Monitoring System, Occupancy Monitoring System and Facial Recognition Access System. The solution is able to operate under complex lighting conditions such as dark light and backlight.
1  Visual Perception

2
Camera Configuration
It adapts to mainstream in-cabin cameras(IR/RGB/RGB-IR)on the market to form an optimized quantitative indicators for image quality to ensure high product consistency on the system level.
2  Camera Configuration

3
Mainstream Chips of the Automotive Cabin Industry
Leveraging on SenseTime’s self-developed HPC and edge computing deployment framework, it utilizes the computing resource of automotive AI chips (NPU/DSP/GPU) targeting the mainstream ECUs in the automotive cabin industry to achieve low latency and less CPU/memory usage.
3  Mainstream Chips of the Automotive Cabin Industry

4
Quality Management System
SenseTime established a comprehensive quality management system, and was granted of the ISO26262 ASIL-B functional safety certification and ASPICE L2 certification in February 2020 and July 2020, respectively.
4  Quality Management System