Scene detection based on video encoding parameters

A method comprising:

• analyzing, by a system including a processor, a video comprising a plurality of sequential frames, including a plurality of intra-frames (I-frames) and a plurality of predictive frames (P-frames);
• identifying, by the system, I-frames that are encoded as I-frames as a result of jitter;
• characterizing, by the system, the I-frames that are encoded as I-frames as the result of jitter as P-frames such that the I-frames that are encoded as I-frames as the result of jitter are not used as boundary points between respective ones of N groups of frames;
• identifying, by the system, the N groups of adjacent frames constituting N scenes of the video based in part on respective locations of the I-frames throughout the plurality of sequential frames, where N is an integer, and using the plurality of I-frames that are not characterized as P-frames as the boundary points between the respective ones of N groups of adjacent frames;
• and
• generating, by the system, one or more video clips from one or more of the N scenes of the video, respectively, using portions of frames respectively associated with the one or more scenes of the N scenes of the video.

A non-transitory computer-readable medium having instructions stored thereon computer-readable instructions that, in response to execution, cause a system including a processor to perform operations, comprising:

• analyzing a video comprising a plurality of sequential frames, including a plurality of intra-frames (I-frames) and a plurality of predictive frames (P-frames);
• identifying I-frames that are encoded as I-frames as a result of jitter;
• treating the I-frames that are encoded as I-frames as the result of jitter as P-frames such that the I-frames that are encoded as I-frames as the result of jitter are not used as boundary points between respective ones of N groups of frames;
• identifying N groups of adjacent frames based in part on respective locations of the I-frames throughout the sequential frames of the video, where N is an integer, and using the plurality of I-frames that are not treated as P-frames as the boundary points between the respective ones of N groups of adjacent frames;
• and
• generating one or more video clips from one or more of the N scenes of the video, respectively, using portions of frames respectively associated with the one or more scenes of the N scenes of the video.