Method and apparatus for calibrating camera

An apparatus for calibrating a camera, comprising:

• at least one processor;
• and
• a memory storing instructions that, when executed by the at least one processor, cause the at least one processor to perform operations, the operations comprising:
• collecting, by the camera and a radar provided on an autonomous vehicle, groups of image-point cloud data, wherein each group of image-point cloud data includes an initial image and point cloud data collected at a same time, the initial image being collected by the camera provided on the autonomous vehicle and the point cloud data being collected by the radar provided on the autonomous vehicle;
• acquiring, from the camera and the radar, an image-point cloud sequence, the image-point cloud sequence including the groups of image-point cloud data;
• determining, after acquiring the image-point cloud sequence, an actual traveling trajectory line of the autonomous vehicle in a three-dimensional space based on the image-point cloud sequence;
• determining, on the actual traveling trajectory line, location points corresponding to the initial images in the groups of image-point cloud data in the image-point cloud sequence to obtain a location point sequence corresponding to the image-point cloud sequence;
• for a location point in the location point sequence, acquiring, at the location point, target point cloud data of a corresponding initial image;
• and
• matching the target point cloud data of the corresponding initial image with the corresponding initial image in the image-point cloud sequence, to determine a correction parameter of the camera,
• wherein the autonomous vehicle performs autonomous driving based on the point cloud data collected by the radar.

A non-transitory computer readable medium, storing a computer program thereon, wherein the computer program, when executed by a processor, causes the processor to perform operations, the operations comprising:

• collecting, by a camera and a radar provided on an autonomous vehicle, groups of image-point cloud data, wherein each group of image-point cloud data includes an initial image and point cloud data collected at a same time, the initial image being collected by the camera provided on the autonomous vehicle and the point cloud data being collected by the radar provided on the autonomous vehicle;
• acquiring, from the camera and radar, an image-point cloud sequence, the image-point cloud sequence including the groups of image-point cloud data;
• determining, after acquiring the image-point cloud sequence, an actual traveling trajectory line of the autonomous vehicle in a three-dimensional space based on the image-point cloud sequence;
• determining, on the actual traveling trajectory line, location points corresponding to the initial images in the groups of image-point cloud data in the image-point cloud sequence to obtain a location point sequence corresponding to the image-point cloud sequence;
• for a location point in the location point sequence, acquiring, at the location point, target point cloud data of a corresponding initial image;
• and
• matching the target point cloud data of the corresponding initial image with the corresponding initial image in the image-point cloud sequence, to determine a correction parameter of the camera,
• wherein the autonomous vehicle performs autonomous driving based on the point cloud data collected by the radar.