Jetvariety is a V4L2 kernel camera driver framework for the Nvidia Jetson platform. It is designed to support Arducam MIPI cameras that are not natively supported by Jetson hardware.

Important Compatibility Notes:

• Do NOT use this driver if you are using an IMX219 or IMX477 based camera module.
• RAW Data Only: Jetvariety primarily supports RAW sensors and provides RAW image data without ISP (Image Signal Processor) support. Users are responsible for implementing software ISP in their own applications. For integrated ISP support, use Arducam OBISP cameras.
• Hardware Requirement: This driver requires a Jetvariety adapter board, which is typically bundled with the camera breakout board.

This driver is part of the Arducam Jetvariety series for Jetson Xavier NX. It only supports specific camera modules. Ensure your hardware matches one of the following SKUs:

• IMX135 (SKU: B0213)
• AR1820 (SKU: B0216)
• AR0230 (SKU: B0247)
• AR1335 (SKU: B0245)
• OV9281 (SKU: B0223)
• OV5647 (SKU: B0215)
• OV2311 (SKU: B0221, B0220, or B0264)

Note for OV2311 users: If the OV2311 fails to retrieve data normally on the first attempt, please try running the application again.

Jetvariety is a V4L2 kernel camera driver framework designed for Nvidia Jetson. It allows users to use Arducam MIPI cameras that are not natively supported by the Jetson platform.

Key benefits:

• Unified Driver: A single-camera driver supports multiple different Arducam cameras, allowing you to switch hardware without changing the driver.
• Software Compatibility: Designed for V4L2 software compatibility on Jetson boards.
• Example: The Arducam_OBISP_MIPI_Camera_Module utilizes this driver.

Using Arducam multi-camera adapter boards, you can connect up to 4 cameras to a single CSI port. On Jetson Nano or Xavier NX development kits (which have two CSI ports), you can connect a maximum of 8 cameras using two adapter boards.

Constraint: Even though multiple cameras are connected to a single CSI port, only one camera can be active at a time. To switch between cameras, you must use GPIO control.

To run the stereo depth demo using Python 2.7, install the required system libraries and Python packages. This setup includes stereovision for stereo processing and matplotlib for visualization.

# Install pip for Python 2.7
wget https://bootstrap.pypa.io/get-pip.py && sudo python get-pip.py

# Install Python packages
sudo pip install stereovision
sudo pip install matplotlib

# Install system dependencies
sudo apt install libpng-dev
sudo apt install libfreetype6-dev
sudo apt install python-gi-cairo

Jetson Nano and Xavier NX provide native support for IMX219 and IMX477 sensors with optimized ISP performance.

• IMX219: The driver and device tree are pre-installed in the official Jetson image. If using an Arducam motorized focus IMX219 module, use the software scripts located in the IMX219 folder to control manual or autofocus functions.
• IMX477: The driver is not pre-installed. You must manually install the driver package from the IMX477 folder to replace the default IMX219 driver. The folder contains packages compatible with different kernel versions.

The PTZ camera control scripts require the smbus library to communicate over I2C. Depending on your Python environment, use the corresponding command:

• For Python 2: sudo apt-get install python-smbus
• For Python 3: sudo apt-get install python3-smbus

The Jetson GPIO Library requires specific user permissions and groups to function correctly. Follow these steps to configure your system:

1. Create a gpio user group and add your current user to it:

   sudo groupadd -f -r gpio
   sudo usermod -a -G gpio $USER

2. Install the custom udev rules by copying the provided rules file to the system directory:
   ```bash
sudo cp /opt/nvidia/jetson-gpio/etc/99-gpio.rules /etc/udev/rules.d/

3. Apply the new rules by reloading udev or rebooting the system:

   sudo udevadm control --reload-rules && sudo udevadm trigger

OR

sudo reboot

For detailed instructions on setting up, installing, and demonstrating Jetvariety cameras on Nvidia Jetson platforms, refer to the official Arducam Camera Demonstration guide.

Note: This README serves as a pointer to the comprehensive documentation hosted externally.

The driver has been moved to releases. For L4T32.4.3 and later versions, it is recommended to use the automatic installation script provided by Arducam.

Refer to the official Arducam documentation for the specific installation command: Automatic driver installation for L4T32.4.3 and later versions.

The driver for Jetson Nano has been moved to the official releases. For L4T (Linux for Tegra) versions 32.4.3 and later, it is recommended to use the automatic installation script provided by Arducam to ensure correct setup.

Detailed instructions for the automatic installation can be found in the Arducam documentation.

https://www.arducam.com/docs/camera-for-jetson-nano/mipi-camera-modules-for-jetson-nano/driver-installation/#2-automatic-driver-installation--only--for-l4t3243-and-after-versions