qr-with-tiling

QR Code Detection with Tiling

This example uses QRDet neural network to detect QR codes. These QR codes are then decoded on the host. The example utilizes tiling to divide the input frame into multiple smaller frames. Each smaller frame is passed to the QR detection network and processed independently. Some sources call this technique SAHI (Slicing Aided Hyper Inference). With it you can detect much smaller objects compared to running inference on the full image but you are running inference multiple times per frame so the FPS is expected to be lower. You can modify this exepriment to detect other classes as well by simply changing the detection model.

NOTE: Due to missing bindings for RVC2, this example only works on OAK4 devices.

Demo

Usage

Running this example requires a Luxonis device connected to your computer. Refer to the documentation to setup your device if you haven't done it already.

You can run the example fully on device (STANDALONE mode) or using your computer as host (PERIPHERAL mode).

Here is a list of all available parameters:

-d DEVICE, --device DEVICE
                    Optional name, DeviceID or IP of the camera to connect to. (default: None)
-fps FPS_LIMIT, --fps_limit FPS_LIMIT
                    FPS limit for the model runtime. (default: 15)
-media MEDIA_PATH, --media_path MEDIA_PATH
                    Path to the media file you aim to run the model on. If not set, the model will run on the camera input.
-r ROWS, --rows ROWS
                    Number of rows in the grid for dividing the output into smaller frames. (default: 2)
-c COLUMNS, --columns COLUMNS
                    Number of columns in the grid for dividing the output into smaller frames. (default: 2)
-is INPUT_SIZE, --input_size INPUT_SIZE
                    Input video stream resolution. {2160p, 1080p, 720p} (default: 1080p)

Peripheral Mode

Installation

You need to first prepare a Python >= 3.10 environment with the following packages installed:

• DepthAI,
• DepthAI Nodes.

Install the zbar library:

sudo apt-get install libzbar0

You can simply install them by running:

pip install -r requirements.txt

Running in peripheral mode requires a host computer and there will be communication between device and host which could affect the overall speed of the app. Below are some examples of how to run the example.

Examples

python3 main.py

This will run the QR Code Detection with Tiling example with the default device and camera input.

python3 main.py -fps 10

This will run the QR Code Detection with Tiling example with the default device at 10 FPS.

python3 main.py -media /path/to/media.mp4

This will run the QR Code Detection with Tiling example with the default device and the specified media file.

python3 main.py -r 3 -c 3

This will run the QR Code Detection with Tiling example with the default device and the specified grid size.

Standalone Mode (RVC4 only)

Running the example in the standalone mode, app runs entirely on the device. To run the example in this mode, first install the oakctl tool using the installation instructions here.

The app can then be run with:

oakctl connect <DEVICE_IP>
oakctl app run .

This will run the example with default argument values. If you want to change these values you need to edit the oakapp.toml file (refer here for more information about this configuration file).