Video processing method and device, and storage medium
AssigneeBeijing Xiaomi Mobile Software Co., Ltd.
InventorsJiagao HU, Fei WANG
Life of this patent
Estimating the term…
Abstract
A video processing method includes: collecting a plurality of video frames sequentially from a to-be-processed video; classifying an image corresponding to each of the plurality of video frames to obtain a classification result; for each video frame, performing a temporal smoothing processing on the classification result for the image corresponding to the video frame, to determine at least one category associated with the video frame; and grouping the adjacent video frames of a same category into a video segment, to obtain a plurality of video segments.
Claims20
A video processing method, comprising:
- collecting a plurality of video frames sequentially from a to-be-processed video;
- for each of the plurality of video frames, classifying an image corresponding to the video frame to obtain a classification result;
- performing a temporal smoothing processing on the classification results for the images corresponding to all of the video frames, to determine at least one category associated with each of the video frames;
- and
- grouping the adjacent video frames of a same category into a video segment, to obtain a plurality of video segments,
- wherein in a case that a same video frame belongs to different categories, the video frame is categorized into different video segments of the categories corresponding to the video frame.
A video processing device, comprising:
- a processor;
- a memory configured to store an instruction executable by the processor;
- and
- a processor is configured to:
- collect a plurality of video frames sequentially from a to-be-processed video;
- for each of the plurality of video frames, classify an image corresponding to the video frame to obtain a classification result;
- perform a temporal smoothing processing on the classification results for the images corresponding to all of the video frames, to determine at least one category associated with each of the video frames;
- and
- group the adjacent video frames of a same category into a video segment, to obtain a plurality of video segments,
- wherein if a same video frame belongs to different categories, the video frame is categorized into different video segments of the categories corresponding to the video frame.
A non-transitory computer-readable storage medium having computer executable instructions stored thereon for execution by a processor to enable the processor to implement a video processing method, the method comprising:
- collecting a plurality of video frames sequentially from a to-be-processed video;
- for each of the plurality of video frames, classifying an image corresponding to the video frame to obtain a classification result;
- performing a temporal smoothing processing on the classification result for the image corresponding to the video frame, to determine at least one category associated with the video frame;
- and
- grouping the adjacent video frames of a same category into a video segment, to obtain a plurality of video segments,
- wherein if a same video frame belongs to different categories, the video frame is categorized into different video segments of the categories corresponding to the video frame.
Dependencies are inferred from the claim text.