System for automatically editing video
View Patent ↗A system provides automated editing of a media file. Frames of a media file are extracted and feature vectors are generated based thereon. The feature vectors are clustered according to similarity of the content of the feature vectors and the temporal proximity of frames corresponding to the feature vectors in the media file. Clusters are ranked and selected according to the ranking. Segments of the media file corresponding to the selected clusters are combined into a highlight video. Hotspots in a media file may be identified in a media file by detecting local maxima in a measure of movement of pixels between frames. Clusters may be ranked and selected according to an iterative algorithm that identifies clusters that are the most different from an average of the feature vectors and from other selected clusters.
1. A method comprising:
receiving, by a computer system, an input media file;
decomposing, by the computer system, the input media file into a sequence frames;
identifying, by the computer system, for each frame subsequent to a first frame of the sequence of frames, motion values of pixels of the each frame relative to a preceding frame of the plurality of frames;
generating, by the computer system, a sequence of filtered motion values by smoothing the motion values of the sequence of frames;
identifying, by the computer system, one or more hotspots in the media file from the sequence of filtered motion values, each hotspot of the one or more hotspots including a video segment of the media file; and
generating an output including the one or more hotspots and excluding portions of the input media file not identified as hotspots.
2. The method of claim 1 , wherein identifying the one or more hotspots in the media file comprises identifying locations in the media file corresponding to one or more local maxima in the sequence of filtered motion values each hotspot of the one or more hotspots including a segment of video corresponding to a local maximum of the one or more local maxima and extending before and after occurrence of the local maximum in the media file.
3. The method of claim 1 , wherein identifying the motion values of the pixels of each frame comprises calculating optical flow of each frame of at least a portion of the sequence of frames.
4. The method of claim 1 , wherein generating the sequence of filtered motion values by smoothing the motion values of the sequence of frames comprises averaging motion values for individual pixels within an individual frame of the sequence of frames.