All News | Boards | Chips | Devices | Software | Archive | About | Contact | Subscribe
Follow LinuxGizmos:
Twitter Facebook Pinterest RSS feed
*   get email updates   *

Low-cost LiDAR camera ships with open source software

Dec 21, 2020 — by Eric Brown 15,926 views

Onion has gone to Crowd Supply to launch a $179, USB-connected “Onion Tau LiDAR Camera” for 3D depth sensing. The 160 x 60-pixel, 81˚ x 30˚ camera offers a open source Tau Studio Web App and Python API software.

After eliciting interest from the Onion community with Onion’s experimental Omega2 LiDAR Kit based on its OpenWrt-on-MIPS based Omega2 module, the company collaborated with Visionary Semiconductor to produce a low-cost 3D LiDAR camera. The result is the $179 Onion Tau LiDAR Camera, which is close to reaching its $15K Crowd Supply goal with 45 days left. Shipments are due May 23.

Onion Tau LiDAR Camera
(click image to enlarge)

Applications for this LiDAR Time of Flight (ToF) camera include augmented reality and computer vision for robotics, automation, self-driving cars, and object and people detection. When combined with an IMU sensor, it also supports environmental mapping applications such as SLAM. The camera offers plug-and-play operation, claims Onion.

The Onion Tau LiDAR Camera provides the same 160 x 60-pixel depth map resolution as Visionary’s $299 Sentinel 3D ToF Camera. According to an Onion comparison chart, this is higher resolution than the $290 Terabee 3Dcam 80×60 and Seeed’s $159 DepthEye 3D cameras, but lower than the $349 Intel RealSense LiDAR Camera L515, at 1024 x 768 (depth) and 1920 x 1080 (2D). Unlike the Terabee and Seeed 3D cameras, as well as Lucid’s over $1,000, 640 x 480, 69° x 51° Helios2 ToF 3D Camera, the new version of its Helios, the Onion Tau also provides a 2D camera with the same 160 x 60-pixel resolution.


The Onion Tau supplies an 81˚ x 30˚ depth and 2D Field of View (FoV) with a range of 0.1 to 4.5 meters. The greyscale, 90 x 41 x 20mm camera supports up to 30fps for depth video and works independently of ambient light.

Onion Tau PCB prototype (left) and Onion Tau Studio app
(click images to enlarge)

The camera connects to any computer via its USB Type-C port and offers 4x M3 mounting holes. There are no details on the components, which are said to enable independent processing of data in real time. Visionary’s Sentinel camera runs on a Cortex-M4-based STM32F446 MCU combined with a ToF sensor.

The Onion Tau LiDAR Camera, which we saw on CNXSoft, is available with open source software including Tau Studio Web App. Based on Python, HTML, and Javascript, the app “takes the realtime depth data coming from the Tau Camera and visualizes it as a greyscale image, a depth map, and a 3D point cloud the user can manipulate – pan, rotate, and zoom,” says Onion. There is also an OpenCV compatible Tau Camera Python library with an API for configuring the camera and capturing depth data.

Further information

The Onion Tau LiDAR Camera is available on Crowd Supply for $179, with shipments from orders made now scheduled for May 23, 2021. More information may be found on Onion’s Crowd Supply page and website.

(advertise here)

Print Friendly, PDF & Email

3 responses to “Low-cost LiDAR camera ships with open source software”

  1. Carlos Alexandre Garcia says:

    Does not exist such a thing as “Lidar Camera”. LiDAR sensors are not cameras. If it’s a camera, it’s not LiDAR. This article is completely wrong about what LiDAR is.

  2. Jeff Child, Chief Editor- LinuxGizmos says:

    Carlos, I’ll grant that the word camera can have both broad and specific meanings. But “Lidar Camera” is definitely a widely used industry term. A traditional camera is simply a sensor of visible light, while an IR camera is a sensor for infrared light. Lidar camera is just as valid a term. All that said, words and meanings should always be questioned, so I thank you for your comment.

  3. Nathan Ma says:

    Hello, Is there a developer here able to use this LiDAR to develop a measurement tools? we hiring developer for this. if there anyone could do it, please contact me, email: [email protected]

Please comment here...