Dynamic audiovisual segment padding for machine learning
Techniques for padding audiovisual clips (for example, audiovisual clips of sporting events) for the purpose of causing the clip to have a predetermined duration so that the padded clip can be evaluated for viewer interest by a machine learning (ML) algorithm. The unpadded clip is padded with audiovisual segment(s) that will cause the padded clip to have a level of viewer interest that it would have if the unpadded clip had been longer. In some embodiments the padded segments are synthetic images generated by a generative adversarial network such that the synthetic images would have the same level of viewer interest (as adjudged by an ML algorithm) as if the unpadded clip had been shot to be longer.
1. A method comprising:
receiving a first unpadded audiovisual segment data set including information indicative of a first unpadded audiovisual segment, with the first unpadded audiovisual segment corresponding to audiovisual data recorded at a sporting contest for a human audience;
determining a set of padding time interval(s) occurring before and/or after the first unpadded segment by subtracting a time length of the unpadded segment from a predetermined minimum time for segment running length that is required to apply a machine learning (ML) algorithm;
for each given padding time interval of the set of padding time interval(s):
determining a respectively corresponding sports viewer interest value that would characterize the given padding time interval if the first unpadded audiovisual segment continued through the given padding time interval and had its sports viewer interest value determined by the machine learning (ML) algorithm, and
generating a padding audiovisual segment for the given padding time interval so that the padding audiovisual segment for the given padding time interval is characterized by the sports viewer interest value determined for the given padding time interval;
assembling the first unpadded audiovisual segment with the padding audiovisual segment(s) corresponding to each padding time interval of the set of padding time interval(s) to obtain a first padded audiovisual segment data set that is sufficiently long in to for application of the ML algorithm to the padded clip; and
training the ML algorithm using the padded clip.
2. A method comprising:
receiving a first unpadded audiovisual segment data set including information indicative of a first unpadded audiovisual segment, with the first unpadded audiovisual segment corresponding to audiovisual data recorded at a sporting contest for a human audience;
determining a set of padding time interval(s) occurring before and/or after the first unpadded segment by subtracting a time length of the unpadded segment from a predetermined minimum time for segment running length that is required to apply a machine learning (ML) algorithm;
for each given padding time interval of the set of padding time interval(s):
determining a respectively corresponding sports viewer interest value that would characterize the given padding time interval if the first unpadded audiovisual segment continued through the given padding time interval and had its sports viewer interest value determined by the machine learning (ML) algorithm, and
selecting, from a repository of historical audiovisual clips, a padding audiovisual segment for the given padding time interval so that the padding audiovisual segment for the given padding time interval is characterized by the sports viewer interest value determined for the given padding time interval;
assembling the first unpadded audiovisual segment with the padding audiovisual segment(s) corresponding to each padding time interval of the set of padding time interval(s) to obtain a first padded audiovisual segment data set that is sufficiently long in to for application of the ML algorithm to the padded clip; and
training the ML algorithm using the padded clip.
3. A method comprising:
receiving a first unpadded audiovisual segment data set including information indicative of a first unpadded audiovisual segment, with the first unpadded audiovisual segment corresponding to audiovisual data recorded at a sporting contest for a human audience;
determining a set of padding time interval(s) occurring before and/or after the first unpadded segment by subtracting a time length of the unpadded segment from a predetermined minimum time for segment running length that is required to apply a machine learning (ML) algorithm;
for each given padding time interval of the set of padding time interval(s):
determining a respectively corresponding sports viewer interest value that would characterize the given padding time interval if the first unpadded audiovisual segment continued through the given padding time interval and had its sports viewer interest value determined by the machine learning (ML) algorithm, and
generating a padding audiovisual segment for the given padding time interval so that the padding audiovisual segment for the given padding time interval is characterized by the sports viewer interest value determined for the given padding time interval;
assembling the first unpadded audiovisual segment with the padding audiovisual segment(s) corresponding to each padding time interval of the set of padding time interval(s) to obtain a first padded audiovisual segment data set that is sufficiently long in to for application of the ML algorithm to the padded clip; and
training the ML algorithm using the padded clip.