HiLens Framework Introduction

The HiLens Framework encapsulates bottom-layer APIs to implement common management functions, enabling you to easily develop skills on the Huawei HiLens management console and cultivate the AI ecosystem.

Figure 1 shows the layered structure of the HiLens Framework. The HiLens Framework encapsulates the bottom-layer multimedia processing libraries (cameras/microphone driver module Media_mini), the image processing library (DVPP) related to D chips, and the model management library (ModelManager). You can also use the familiar visual processing library OpenCV. In addition, the HiLens Framework provides the following modules for you to develop skills, such as human figure detection, facial recognition, and fatigue driving detection. Table 1 details the modules.

Figure 1 HiLens Framework
Click to enlarge

**Table 1** Module description
No.	Module	Function
1	Input Manager	Input module: manages the access of input data such as video and audio data.
2	Media Processor	Preprocessing module: processes media data such as video and audio data.
3	Model Manager	Model management module: initializes models and infers based on the models.
4	Output Manager	Output module: manages output tasks such as streams, files, and message notifications.
5	Resource Manager	Resource management module: manages paths of resources such as files, images, and models.
6	Logging System	Log module: manages the log system.