Follow-up shooting method and device, medium and electronic device

1. A follow-up shooting method comprising:

acquiring a video image in a live classroom, wherein the video image comprises at least one object;

analyzing the video image to obtain a gesture image for controlling a camera; and

determining whether to control the camera to track and shoot an execution object related to the gesture image based on the gesture image,

wherein analyzing the video image to obtain a gesture image for controlling a camera comprises:

analyzing the video image to obtain a hand image of the at least one object;

determining positions of bone key points and marks of bone key points of a hand based on the hand image;

connecting the positions of bone key points corresponding to the marks of bone key points based on a marking rule of bone key points to obtain a hand bone image; and

determining that the gesture image is presented as the gesture bone image in response to the hand bone image matching a gesture bone image, wherein the gesture bone image is preset.

2. The method according to claim 1 , wherein,

the gesture bone image comprises a first gesture bone image or a second gesture bone image;

determining whether to control the camera to track and shoot an execution object related to the gesture image based on the gesture image comprises:

determining to control the camera to perform a panoramic shooting based on specific panoramic parameters in response to the gesture image being represented as the first gesture bone image;

determining to control the camera to track and shoot the execution object related to the gesture image in response to the gesture image being represented as the second gesture bone image.

3. The method according to claim 2 , wherein the method further comprises:

acquiring current shooting parameters after determining to control the camera to track and shoot the execution object related to the gesture image;

analyzing the video image to obtain a body image of the execution object related to the gesture image;

obtaining shooting adjustment parameters for tracking the body image according to the video image and the body image on the basis of the current shooting parameters;

adjusting the camera to track and shoot the execution object based on the shooting adjustment parameters.

4. The method according to claim 3 , wherein obtaining shooting adjustment parameters for tracking the body image according to the video image and the body image on the basis of the current shooting parameters comprises:

acquiring a target core position of the execution object in the video image based on the body image of the execution object, where the target core position is a focus position of the camera for tracking and shooting;

obtaining the shooting adjustment parameter for tracking the target core position according to the video image and the target core position on the basis of the current shooting parameters.

5. The method according to claim 4 , wherein obtaining shooting adjustment parameters for tracking the body image according to the video image and the body image on the basis of the current shooting parameters further comprises:

acquiring a specific point position in the video image;

obtaining the shooting adjustment parameters of the execution object according to the target core position and the specific point position on the basis of the current shooting parameters, wherein the shooting adjustment parameters enable the specific point position of the video image captured after adjustment to coincide with the target core position of the video image.

6. The method according to claim 4 , wherein the target core position comprises: a center position of a head image of the execution object, a geometric center position of the body image of the execution object, or a center point of the line connecting the two furthest points in the body image.

7. A non-transitory computer-readable storage medium with a computer program stored thereon, wherein, when the program is executed by a processor, the method according to claim 1 is implemented.

8. An electronic device, comprising:

one or more processors;

a storage to store one or more programs,

wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to claim 1 .

9. A follow-up shooting device, comprising:

an acquirer to acquiring a video image in a live classroom, wherein the video image comprises at least one object;

an analyzer to analyze the video image to obtain a gesture image for controlling a camera; and

a determiner to determining whether to control the camera to track and shoot an execution object related to the gesture image based on the gesture image,

wherein the analyzer analyzes the video image to obtain a hand image of the at least one object; determines positions of bone key points and marks of bone key points of a hand based on the hand image; connects the positions of bone key points corresponding to the marks of bone key points based on a marking rule of bone key points to obtain a hand bone image; and determines that the gesture image is presented as the gesture bone image in response to the hand bone image matching a gesture bone image, wherein the gesture bone image is preset.