Information generation method and apparatus

Abstract

The embodiment of the invention discloses an information generation method and apparatus. A specific implementation manner of the method comprises the steps of obtaining a to-be-detected image, and first area position information used for defining a human face area in a previous one frame of image of the to-be-detected image; based on the first area position information, performing a human face key point positioning operation on the to-be-detected image to obtain a coordinate of each human face key point included in the to-be-detected image; based on the coordinate of the each human face key point, determining second area position information used for defining the human face area in the to-be-detected image; and generating corresponding relationship information between the to-be-detected image and the second area position information. According to the implementation manner, the information generation rich in pertinence is implemented.

Description

technical field [0001] The present application relates to the field of computer technology, specifically to the field of Internet technology, and in particular to an information generation method and device. Background technique [0002] Existing tracking methods for targets (such as target faces, etc.) are usually correlation-based tracking methods. For example, through some correlation judgment methods, the targets of different frames are connected in series to obtain the target tracking information. But the existing object tracking methods have the problem of poor robustness. For example, when a non-target and a target are occluded, it often causes the temporary disappearance of the target. Or when the target approaches or moves away from the camera quickly, the size of the target may suddenly increase or decrease, which is easy to lose. When the target appears again, it is often impossible to continue to track the target accurately. Contents of the invention [000...

AI Technical Summary

Problems solved by technology

However, the existing target tracking methods have the problem of poor robustness

For example, when non-targets and targets are occluded, it often leads to temporary disappearance of targets

Or when the target approaches or moves away from the camera quickly, it may cause the target size to increase or decrease suddenly, which is easy to lose

When the target appears again, it is often impossible to continue to track the target accurately

Images