| CPC G06T 3/4023 (2013.01) [G06T 5/73 (2024.01); G06T 2207/10016 (2013.01)] | 16 Claims |

|
1. A video processing method, comprising:
determining a blurred video frame in an initial video;
removing the blurred video frame from the initial video to obtain an intermediate video which does not comprise the blurred video frame;
determining a video frame to be inserted, whose timestamp is a target timestamp in the intermediate video, based on video frames whose timestamps are adjacent to the target timestamp, wherein the target timestamp is a timestamp of the blurred video frame; and
inserting the video frame to be inserted at a position in the intermediate video corresponding to the target timestamp to obtain a target video;
wherein determining the blurred video frame in the initial video comprises:
determining a first edge image at a preset position in a current video frame based on Sobel operator, wherein the current video frame is any video frame in the initial video; and
determining the blurred video frame based on an edge image at a preset position in each video frame in the initial video;
wherein determining the blurred video frame based on the edge image at the preset positions in each video frame in the initial video comprises:
determining a first cumulative sum of pixel values of various pixels in the first edge image;
determining a second cumulative sum of pixel values of various pixels in a second edge image at a preset position of a neighboring video frame, wherein the neighboring video frame is a video frame adjacent to the current video frame in the initial video;
determining an absolute value of a first difference between the first cumulative sum and the second cumulative sum; and
determining the blurred video frame based on the absolute value of the first difference.
|