Experience AI Applications with Qdemo UI - Qualcomm Dragonwing Documentation

Currently, the Qdemo application isn’t supported through the GUI launcher icon. Launch the Qdemo application using the device terminal. QDemo is available only for server builds.

Qualcomm Dragonwing™ platform come preinstalled with Qdemo, a graphical user interface (GUI) application designed to run AI use cases. Use the Qdemo application to explore the AI capabilities of the development kit. The following table lists the AI sample applications that you can run using Qdemo. Sample applications supported in Qdemo

Application	Description
Daisychain Pose	Performs cascaded object detection and pose detection streamed from a camera.
FaceDetection	Processes inputs from a camera and uses the Qualcomm® AI Hub detection model for face
Multistream	Shows AI inference (object detection) on input streams from a camera or a file.
ObjectDetection	Performs object detection on streams from a camera.
Parallel-AI-Fusion	Performs object detection, image classification, pose detection, and image segmentation on an input stream from a camera.
DashCamera	Displays two simultaneous camera feeds side by side.
Record live video	Records live video from a camera source, simultaneously displaying the stream and saving it as an H.264-encoded MP4 file (video_out.mp4).
VideoTestSrc	Renders a GStreamer test pattern using videotestsrc, useful for verifying the display pipeline without a physical camera.
VideoWall	Plays a previously recorded MP4 file using the Record live video application across a 2×2 tile grid for a 4-up video wall display.

For the complete list of available applications, see Qualcomm IM SDK sample applications.

Prerequisites

Ubuntu multimedia image: The device must be flashed with the prebuilt ubuntu image, which includes Qdemo and the Qualcomm IM SDK.
HDMI monitor: Connect to the device to view Qdemo output.
Mouse and keyboard: Required to interact with the Weston desktop and the Qdemo application.
Wi-Fi connection: Required for AI sample applications to download models and labels from GitHub. Not required for multimedia applications.
USB camera (optional): Required for camera-based sample applications. If connected, select it from the Source list in Qdemo.

Run sample applications using Qdemo

Connect the mouse, keyboard, and HDMI monitor, then power on the device

Connect the mouse, keyboard, and HDMI monitor to the device, then power it on.

Open the target device shell

ssh ubuntu@<IP address of the target device>

Install the required dependencies

sudo apt-get install -y libgstreamer1.0-dev gstreamer1.0-libav gstreamer1.0-alsa gstreamer1.0-gtk3

sudo apt-get install -y python3-gi-cairo gir1.2-gtk-3.0

Create the required folders and set ownership

Create media, labels, models, and a data folder, and change owner to ubuntu:ubuntu:

sudo mkdir -p /etc/media/ /etc/labels/ /etc/models/ /etc/data/

sudo chown -R ubuntu:ubuntu /etc/media/ /etc/labels/ /etc/models/ /etc/data/

Download the required artifacts

wget -P /etc/media/ https://raw.githubusercontent.com/qualcomm/sample-apps-for-qualcomm-linux/refs/heads/main/scripts/download_artifacts.sh

wget -P /etc/media/ https://raw.githubusercontent.com/qualcomm/sample-apps-for-qualcomm-linux/refs/heads/main/qualcomm-linux/artifacts/qdemo/Qdemo.gif

wget -P /etc/media/ https://raw.githubusercontent.com/qualcomm/sample-apps-for-qualcomm-linux/refs/heads/main/qualcomm-linux/artifacts/qdemo/Qdemo.png

Launch the Qdemo application from the device terminal

gst-gui-launcher-app.py

The Experience Application Demo window will pop-up.

Select Wi-Fi to connect to a network

Only AI sample applications require a Wi-Fi connection. For multimedia applications, skip this step.

Wi-Fi selection in the Qdemo application

Select the required camera in the Source list

Not all applications require a camera. If a USB camera is connected, select it from the list. The default selection is On-Device-Camera.

Select an application and click Start

In the Applications list, select an application and click Start.The application launches and displays output on the screen.

During the first run, the application may take more than a minute to start, as it downloads AI models and labels from GitHub.

The following is the sample output of the ObjectDetection application:

Sample output of the ObjectDetection application in Qdemo

Close the output window to stop the sample application

Close the output window to stop the sample application.

Select Exit to close the Qdemo application

Select Exit to close the Qdemo application.

​Prerequisites

​Run sample applications using Qdemo

Prerequisites

Run sample applications using Qdemo