LIPSedge M3 Embedded 3D ToF Camera w/ LED & IMU

The camera is compatible with Windows and various Linux distributions. For software development, it supports the OpenNI framework for depth data streaming and OpenCV for computer vision processing. Drivers and SDK samples are typically provided by LIPS to interface with the USB 2.0 device, enabling rapid application development on platforms like NVIDIA Jetson, Intel NUC, and Raspberry Pi. You'll need to install the vendor-specific libraries before accessing the depth stream and IMU data in your code.

The Near Mode optimizes depth accuracy for objects very close to the sensor, ranging from 0.2 to 1.0 meters, ideal for detailed gesture recognition or precise pick-and-place operations. The Normal Mode extends the operational range from 1 to 4 meters, suitable for broader scene mapping, people counting, or navigation. You select the mode via the provided API or configuration tool based on your application's required working distance and the level of detail needed for depth data processing.

The integrated accelerometer and gyroscope data is streamed alongside the depth frames through the camera's USB interface. Using the provided SDK, you can read the raw or filtered IMU data (acceleration and angular rate) to implement features like gesture recognition, tilt compensation for the depth map, or inertial navigation in robotics. The sensor fusion between depth and motion data typically requires custom application logic, but the SDK provides the necessary hooks to synchronize and timestamp both data streams.

The module requires a 5V DC supply and consumes approximately 4W (around 800mA). A standard Raspberry Pi 4 USB port can supply up to 1.2A at 5V, which is generally sufficient. However, for stable operation, especially when other peripherals are connected, it is recommended to use a powered USB hub or a dedicated 5V/1A power adapter connected via the camera's Micro-B port to avoid potential brown-outs or data corruption.