A follow-up control method, device, computer equipment and storage medium

Abstract

The invention discloses a follow-up control method, device, computer equipment and storage medium. The method includes: obtaining a horizontal offset between the horizontal center of the target bounding box and the horizontal center of the screen and a vertical offset between the vertical center of the target bounding box and the vertical center of the screen; according to the horizontal offset and The vertical offset calculates the horizontal movement amount and the vertical movement amount of the pan-tilt; the movement of the pan-tilt is controlled according to the horizontal movement amount and the vertical movement amount. The invention controls the movement of the pan-tilt according to the bounding box of the target, so as to realize the automatic follow-up shooting of the target and achieve a good composition effect.

Description

technical field [0001] The present invention relates to the field of camera technology, and in particular, to a tracking control method, device, computer equipment and storage medium. Background technique [0002] With the widespread application of deep learning technology, it has become a no-brainer to obtain portrait bounding boxes or face bounding boxes from captured images. However, how to adjust the rotation angle of the gimbal to follow the target object according to the bounding box is still a difficult problem. [0003] In the prior art, the face bounding box is directly placed at the center (or the upper center) of the entire picture, but this control method has obvious disadvantages: first, it is not sensitive enough to the action of the shooting object, and when the shooting object is not sensitive enough, When the limbs move, because the position of the face does not change and the shooting picture does not change, the movements of the limbs may exceed the shoot...

AI Technical Summary

Problems solved by technology

However, how to adjust the rotation angle of the gimbal according to the bounding box to follow up the target object is still a difficult problem

[0003] In the prior art, the face bounding box is directly placed at the center (or above the center) of the entire picture, but this control method has obvious disadvantages: first, it is not sensitive enough to the motion of the subject, when the subject When the limbs move, because the position of the face does not change, the shooting picture remains unchanged, so the movement of the limbs may exceed the shooting picture; the second is to restrict the movement of the subject, when the subject turns around, because the face cannot be observed First, the shooting equipment can only keep the original shooting angle unchanged, so the equipment requires that the subject cannot be in the state of facing away for a long time, and if it is in the state of facing away, it cannot move in a large range; the third is that there is a certain difficulty in estimating the size of the subject , when the subject is sideways, the bounding box of the face is deformed, and it is difficult for the shooting equipment to accurately estimate the scale of the real subject at this time, so it is difficult to set a reasonable focal length value

Images