Systems and methods for determining activities performed during video capture
Systems and method of determining one or more activities performed during video capture are presented herein. Information defining a video may be obtained. The information defining the video may include content information, metadata information, and/or other information. The content information may define visual content of the video and/or other content of the video. The metadata information may include tracking information and/or other information. The tracking information may locations of the capture device as a function of progress through the video. One or more activity types being performed by an entity moving with the capture device during the previous capture may be determined based on the tracking information. For example, a location graph may be generated from the tracking information. The location graph may be used to determine one or more activity types that were performed.
1. A system configured to recognize one or more activities performed during video capture, the system comprising:
one or more physical processors configured by machine-readable instructions to:
obtain information defining a video, the video including visual content previously captured by a capture device over a capture duration, the information defining the video including content information and metadata information, the content information defining the visual content, the metadata information including tracking information, the tracking information including locations of the capture device as a function of progress through the capture duration, the locations including a first set of locations; and
determine a type of activity being performed by an entity moving with the capture device over the capture duration based on the tracking information, wherein the determining the type of activity being performed by the entity moving with the capture device over the capture duration based on the tracking information comprises:
generating a location graph for the video using the tracking information the location graph including the first set of locations of the capture device as the function of the progress through the capture duration;
identifying an overall shape of the location graph; and
determining the type of activity being performed by the entity moving with the capture device based on the overall shape of the location graph, wherein determining the type of activity includes distinguishing between a sport-type activity and a race-type activity based on the overall shape of the location graph.
2. The system of claim 1 , wherein the type of activity is determined to be a sport-type activity based on the overall shape of the location graph matching a shape of an athlete-specific movement.
3. The system of claim 1 , wherein the location graph is a two-dimensional graph characterizing the locations of the capture device with respect to two coordinate directions of movement.
4. The system of claim 1 , wherein the type of activity is determined to be a race-type activity based on the overall shape of the location graph matching a shape of a track.
5. The system of claim 1 , wherein the determining the type of activity being performed by the entity moving with the capture device over the capture duration based on the tracking information further comprises:
identifying a speed of the entity moving with the capture device; and
determining a sub-type of the type of activity being performed by the entity moving with the capture device based on the speed of the entity.
6. The system of claim 5 , wherein different sub-types correspond to different types of vehicles or equipment used by the entity.
7. The system of claim 1 , wherein the metadata information further includes timing information, the timing information characterizing the tracking information as the function of the progress through the capture duration.
8. The system of claim 1 , wherein the video further includes audio content, and wherein the content information further defines the audio content of the video.
9. The system of claim 1 , wherein a trained neural network is utilized to determine the type of activity.
10. The system of claim 1 , wherein the progress through the capture duration is represented as one or more of real-time progress, reduced-time progress, or speeded-up progress.
11. A method of recognizing one or more activities performed during video capture, the method being implemented by a computer system comprising one or more physical processors and non-transitory storage media storing machine-readable instructions, the method comprising:
obtaining information defining a video, the video including visual content previously captured by a capture device over a capture duration, the information defining the video including content information and metadata information, the content information defining the visual content, the metadata information including tracking information, the tracking information including locations of the capture device as a function of progress through the capture duration, the locations including a first set of locations; and
determining a type of activity being performed by an entity moving with the capture device over the capture duration based on the tracking information, wherein the determining the type of activity being performed by the entity moving with the capture device over the capture duration based on the tracking information comprises:
generating a location graph for the video using the tracking information, the location graph including the first set of locations of the capture device as the function of the progress through the capture duration;
identifying an overall shape of the location graph; and
determining the type of activity being performed by the entity moving with the capture device based on the overall shape of the location graph, wherein determining the type of activity includes distinguishing between a sport-type activity and a race-type activity based on the overall shape of the location graph.
12. The method of claim 11 , wherein the type of activity is determined to be a sport-type activity based on the overall shape of the location graph matching a shape of an athlete-specific movement.
13. The method of claim 11 , wherein the location graph is a two-dimensional graph characterizing the locations of the capture device with respect to two coordinate directions of movement.
14. The method of claim 11 , wherein the type of activity is determined to be a race-type activity based on the overall shape of the location graph matching a shape of a track.
15. The method of claim 11 , wherein the determining the type of activity being performed by the entity moving with the capture device over the capture duration based on the tracking information further comprises:
identifying a speed of the entity moving with the capture device; and
determining a sub-type of the type of activity being performed by the entity moving with the capture device based on the speed of the entity.
16. The method of claim 15 , wherein different sub-types correspond to different types of vehicles or equipment used by the entity.
17. The method of claim 11 , wherein the metadata information further includes timing information, the timing information characterizing the tracking information as the function of the progress through the capture duration.
18. The method of claim 11 , wherein the video further includes audio content, and wherein the content information further defines the audio content of the video.
19. The method of claim 11 , wherein a trained neural network is utilized to determine the type of activity.
20. The method of claim 11 , wherein the progress through the capture duration is represented as one or more of real-time progress, reduced-time progress, or speeded-up progress.