pdf-icon

StackFlow AI Platform

Module LLM Applications

CV Vision Application

Vision Language Model (VLM)

Large Language Model (LLM)

Voice Assistant

AI Pyramid - Box Demo

BoxDemo is a high-performance multi-channel video detection demo program officially provided by AXERA. It supports simultaneous encoding/decoding and object detection for up to 16 video streams. The program can detect objects such as motor vehicles, non-motor vehicles, pedestrians, and license plates, and outputs real-time detection results via the Display #0 interface. AI Pyramid can be used as an efficient multi-channel video processing computing box.

1. Obtain Test Videos

You can obtain the test videos using one of the following methods:

Method 1: Manual Download and Upload

Visit the BoxDemo Test Video Collection to download the files, then upload them to the AI Pyramid device.

Method 2: Command-Line Download

wget https://m5stack.oss-cn-shenzhen.aliyuncs.com/resource/linux/AI_Pyramid/boxvideos.zip
Dependency Check
If the zip tool is not installed on the system, please run apt install zip to install it first.

2. Deploy Files

Unzip the video package and replace the corresponding configuration files:

unzip boxvideos.zip
rm boxvideos.zip
cp boxvideos/run.sh /opt/bin/BoxDemo/
cp boxvideos/box.conf /opt/bin/BoxDemo/

3. Run BoxDemo

Execute the startup script to run the program:

/opt/bin/BoxDemo/run.sh
Resource Conflict Notice
If other programs have already loaded and occupied AI model resources, BoxDemo initialization may fail. Please ensure that system resources are sufficient and that no other AI applications are running simultaneously.

After the program starts, please wait patiently for it to finish loading. Once loading is complete, the Display #0 interface will output real-time multi-channel detection video and detection results.

On This Page