RGBD ToF Depth Camera Module

Overview¶

The Arducam RGBD ToF Camera Kit (also known as Arducam T2 All-in-One Kit) is the ultimate turnkey solution for developers needing immediate access to high-precision depth sensing. Integrating a 16MP RGB sensor and a VGA ToF depth sensor, this kit delivers 1% depth accuracy and frame rates up to 55fps on the Raspberry Pi 5.

Whether for robotics, AR/VR, or industrial automation, this kit streamlines the path from prototyping to production.

Left: All-in-One Kit | Right: Camera Module Only

Key Applications¶

Versatile Use Cases

Robotics: Obstacle avoidance, SLAM navigation, and object recognition.
Augmented Reality (AR) / VR: High-precision environmental depth sensing.
3D Scanning: Detailed 3D model reconstruction.
Industrial Automation: Quality control, package sorting, and volume measurement.
Smart Home: Gesture control and security monitoring.
Medical Devices: Non-invasive depth measurement for diagnostics.

Specifications¶

Comparison: Kit vs. Module¶

Feature	All-in-One Kit (B0548)	Camera Module (B0556)
Image
Component Type	Complete Solution (w/ Pi 5 inside)	Integration Module
Resolution	16MP (RGB) + VGA (ToF)	16MP (RGB) + VGA (ToF)
Stream Type	RGB + Depth Map	RGB + Depth Map
Interface	MIPI CSI-2	MIPI CSI-2
Target Use	Quick Prototyping / Turnkey	Custom Integration

Hardware Setup¶

Device List¶

Arducam RGBD ToF Camera Module(B0556) * 1
2-pin Dupont Cable * 1
22-22pin FFC Cable * 2
Platform(Raspberry Pi 5, etc.)

Installation Guide¶

Connect the RGBD ToF Camera Module using two 22-22 Pin FFC Cables.

Connect the 2-pin DuPont wire to the RGBD ToF Camera Module.

Connect the Dupont cables to the corresponding interfaces (one-to-one correspondence, do not connect haphazardly):

Connect the red cable to pin 4 (5V) and the black cable to pin 6 (GND).

As shown below, connect the ToF Depth interface to the cam1 port of the Raspberry Pi, and connect the RGB interface to the Raspberry Pi cam0 port.

Critical Warning: Do Not Disassemble

Strictly NO Disassembly for the RGB Depth Camera Module (B0556).

Do not remove the screws marked 1 and 2 in the diagram below. The module is calibrated at the factory; unauthorized disassembly will cause the calibration to deviate.

Final Connection Diagram

Software Guide¶

Image Flashing¶

Download the Image¶

Arducam RGBD Camera Pre-built Image

Flash the Image¶

Option I

Download the tool: https://www.raspberrypi.com/software/

Operation Guide: getting-started raspberry-pi-imager

Option II

Flash an image RPI Tutorial: Raspberry Pi Tutorial - flash-compute-module-emmc

Pre-built Demonstrations¶

1. ToF Only (Depth)2. RGB-D (Fusion)3. Advanced Apps

Access raw depth data without the RGB overlay.

1. Depth Preview (run-tof)

Preview the depth map and infrared intensity.

Close Range Mode (Near):
```
run-tof --mode 0
```
Long Range Mode (Far):
```
run-tof --mode 1
```

ToF Only Visualization — From Top to Bottom: Confidence, Depth, Greyscale

2. ToF Point Cloud (run-tof-pointcloud)

Visualize the raw 3D depth cloud (grayscale based on intensity).

run-tof-pointcloud

ToF Point Cloud — ToF-only Point Cloud View

Combine the 16MP Camera color data with the ToF depth sensor.

3. RGBD Capture (run-rgbd)

Displays the color video stream synchronized with the depth map.

run-rgbd

RGBD Capture Demo — Aligned RGB and Depth View

4. RGBD Point Cloud (run-rgbd-pointcloud)

Renders a real-time, fully colored 3D Point Cloud.

run-rgbd-pointcloud

RGBD Point Cloud — RGB-Mapped 3D Point Cloud

5. Volume Measurement (run-rgbd-volume)

Dynamically calculates the volume of objects using depth contours.

run-rgbd-volume


Mixed Window Displays object volume info.	Contours Window Displays the selected object.	Mask Window White: Reference Plane Gray: Object

6. ROS 2 Integration (ros demo)

Launch the standard ROS2 node and visualization tools.

Step 1: Start the Node (on Raspberry Pi)

ros-setup
ros2 run arducam_rclcpp_rgbd_pointcloud rgbd_pointcloud

Step 2: Visualization

Option A: Local Visualization

Run directly on the Raspberry Pi(in a new terminal):
```
ros-setup
rviz2
```
Local Rviz2 Visualization
Option B: Remote LAN Visualization

You can also observe the data on another PC connected to the same Local Area Network (LAN) by launching rviz2 on the remote machine.

Remote Visualization via LAN

Developer Compilation Guide¶

If you intend to modify the source code or integrate the driver into your own C++ project, follow the compilation instructions below.

1. ToF Only (Depth)2. RGBD (Fusion)

Use these instructions to compile the standalone depth sensor applications.

1. Basic Example

Compile the demo script:

cd ~/tof-examples
bash ./compile.sh

Success Indicator

If compilation is successful, the terminal will verify the path:
== Run /home/pi/tof-examples/build/example/cpp/example

Run the compiled binary:

/home/pi/tof-examples/build/example/cpp/example

2. Point Cloud

Compile the point cloud demo:

cd ~/tof-examples
bash ./compile_pointcloud.sh

Success Indicator

If compilation is successful, the terminal will verify the path:
== Run /home/pi/tof-examples/build/open3d_preview/preview_pointcloud

Run the compiled binary:

/home/pi/tof-examples/build/open3d_preview/preview_pointcloud

Use these instructions to compile the combined RGB + ToF applications.

Restriction Note

Please note that the source code for the example and volume_measurement demos is not currently supported for user compilation.

1. Capture Demo

Compile the source code:

cd ~/rgbd-examples/capture
cmake -Bbuild && cmake --build build -j4

Run the application:

# Run C++ Version
/home/pi/rgbd-examples/capture/build/capture

# Run C Version
/home/pi/rgbd-examples/capture/build/capture_c

Tip

Press q in the preview window to exit the program.