Video filtering method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a video filtering method and device, electronic equipment and a storage medium. The video filtering method comprises the steps of extracting a target area of each frame of image in a to-be-filtered video through a pre-trained image segmentation network; performing dynamic background modeling on each frame of image in the to-be-filtered video to obtain a general background image of the to-be-filtered video, wherein the general background image does not include the to-be-filtered object; extracting a target background image corresponding to the target area from the universal background image, wherein the target background image and the target area have the same shape; and replacing the target area with the target background image so as to filter the target area. By replacing the target area with the target background image corresponding to the target area in the universal background image, the content of the original target area is not displayed any more, and no new element is added, and the attention of the user is not distracted, so that the user experience is improved.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a video filtering method, device, electronic equipment and storage medium. Background technique [0002] Artificial intelligence is widely used in the field of security, by deploying cameras in various demand scenes for video surveillance of the demand scene. In some video surveillance scenarios, the target object will be captured. For privacy protection or other purposes, it is necessary to protect the captured target object so that some observers without authorization cannot identify who is captured. Thereby protecting the privacy of the person being photographed. The traditional protection process is to blur the detected target object. For example, after manually marking the target object or performing target detection, blur processing is performed in a large range to make the characteristics of the target object indistinguishable. However, due to the blurred ...

AI Technical Summary

Problems solved by technology

The traditional protection processing is to blur the detected target object. For example, after manually marking the target object or performing target detection, blur processing is performed in a large range to make the characteristics of the target object indistinguishable. However, due to the blurred The target object still exists in the video image, and, for the blurred object, the observer of the video tends to allocate more attention to want to distinguish the blurred object, so that the observer's attention is distracted , resulting in a degraded user experience

Images