[Video] Extend Your Vision: VisLab’s 3DV Stereo System 扩展你的视觉:VisLab的立体视觉系统

3DV, the three dimensional vision, developed by Vislab srl, Parma, Italy, is a stereoscopic system capable of providing a dense 3D map of the surroundings and a quick interpretation of the environment, detecting and tracking obstacles and estimating terrain surface, and slope.

Being developed by Vislab throughout the years, now 3DV has the ability to provide up to 3 million (or 7 million, depends on the specific version) 3D point cloud per second in real time. 3DV has a wide range of applications, especially for robotics and vehicular platforms.

Comparing to laser, like the famous 3D laser by Velodnye, camera-based 3DV has shown lots of its advantages. 3DV is a low-cost, low-power, compact and rugged intelligent sensor, while it can also provide abundant information. The basic version provides the 3D depth map and raw images, additional plugins are available to offer enhanced perception capabilities, including: obstacle detection, terrain mapping, pedestrian classification, vehicle classification, visual odometry, and lane detection. In addition, 3DV also supports the following frameworks: ROS, RTMaps, ADTF, and Halcon. These amazing features enhances the perception tasks for roboticists and engineers in various field.

Besides, 3DV is very easy to install and use. It offers the user-friendly interfaces and an open source cross-platform (Linux, Windows) C++ library to interface with the camera, and an open source cross-platform client to set camera properties, show, and dump images.

Below is the recorded webinar hosted by AutonomouStuff on “3DV: A Tool for World Reconstruction through Artificial Vision”. The webinar is given by president of VisLab srl, Professor Alberto Broggi, and researcher, Paolo Zani.

There is also another open introduction 3DV-E An Embedded, Dense Stereovision-based Depth Mapping Device, given by one of the project managers, Luca Bombini, from VisLab srl.

Until now, 3DV has two different versions. Version A, the 3DV-A uses an unboxed board as its processing engine. Version E, the 3DV-E, is the embedded version of 3DV, based on an FPGA. Below are technical details about 3DV-A and 3DV-E, respectively.




If you are interested in the 3DV system, please refer to the official website of 3DV-A and 3DV-E to order. You can also contact AutonomouStuff’s page for stereo vision products for details. The Japanese page on 3DV is here, by ZMP. Until now there is no Chinese agency for 3DV. If any company wants to agent the 3DV in China, I’d like to offer you my help contacting VisLab.

Images and videos are mainly based on the official website of Vislab, including the pages/YouTube channel/datasheets, .etc.


无人驾驶汽车研究员 半吊子摇滚混子乐手 Researcher on Intelligent Vehicle Guitarist/Singer/Rocker
此条目发表在Autonomous Driving分类目录,贴了, , , , , , 标签。将固定链接加入收藏夹。