With the development of IoT, the demand for 3D technology has become more and more. A large number of recognition no longer depends on people, but robots are required to perform autonomous object recognition. How to empower 3D recognition functions on robots has become very important.
In the past four or five years, the surrounding environment and experience have been empowered by 3D perception, such as 3D human faces, autonomous driving, robot vacuum, logistics sorting, furniture design, somatosensory games, etc. Unconsciously, the surrounding area is surrounded by a large number of 3D perception technologies. Although 3D perception technologies have not been fully commercialized, they are getting faster and faster. Some technologies have been used in the industry and mass-produced, especially in the robotics industry. Robots need to be intelligent, and robot movement and obstacle avoidance require intelligent 3D perception technology.
DOMI focuses on the development and provision of intelligent sensing and artificial intelligence system products and solutions. At present, there are three mainstream solutions for 3D vision: binocular, structured light, and TOF. Binocular is the most mature and earliest; structured light is recognized by the industry in terms of accuracy; TOF camera also has its own unique features, mainly to perceive 3D technology more directly, and it has some advantages, the volume can be made very small, Costs are also getting cheaper. However, the TOF camera module is very complex. Although the sparrow is small and complete, it has a lot of content, including receivers and transmitters. Because it is an active light, it requires certain correction and algorithm processing for the depth test of lidar. The entire TOF The camera system is also quite complex. Our TOF camera system is constantly evolving. The first is the integration of the system. The TOF camera system we provide is not a simple chip solution for customers. We integrate the reception, transmission and processor of the entire TOF camera system as a whole. It can be well placed on the robot and achieve better results.
