| abstract |
The invention relates to a virtual point cloud three-dimensional target detection method based on supervised monocular depth estimation. Depth estimation model. The beneficial effects of the present invention are: the present invention directly uses the camera as the main sensing means, avoids the application of expensive sensors such as lidar in the three-dimensional target detection system, and also directly avoids the joint calibration existing in the multi-sensor sensing method. The problem of synchronization with data further reduces the cost of sensor deployment, reduces the dependence on lidar, helps reduce the cost of 3D target detection methods, and promotes the application of this technology in various fields. In addition, the algorithm model is deployed to edge devices through offline training and online prediction, which relieves the computing pressure of devices and improves the intelligence level of edge devices. |