Methods and systems of spatiotemporal pattern recognition for video content development
Presenting event-specific video content that conforms to a user selection of an event type includes processing at least one video feed through at least one spatiotemporal pattern recognition algorithm that uses machine learning to develop an understanding of at least one event within the at least one video feed to determine at least one event type, wherein the at least one event type includes an entry in a relationship library at least detailing a relationship between two visible features of the at least one video feed, extracting the video content displaying the at least one event and associating the understanding with the video content in a video content data structure. A user interface is configured to permit a user to indicate a preference for at least one event type that is used to retrieve and provide corresponding extracted video content with the data structure in a new video feed.
1. A method for animating motion of a second player in a video of a first player, comprising:
processing at least one video feed of the first player through at least one spatiotemporal pattern recognition algorithm that is trained through a use of a machine learning system to detect at least one first player event within the at least one video feed, the training based on adjusting an input feature and a weight within the machine learning system, wherein the input is selected from the group consisting of:
relative direction of motion of at least two visible features,
duration of relative motion of visible features with respect to each other,
rate of motion of at least two visible features with respect to each other,
acceleration of motion of at least two visible features with respect to each other, and
projected point of intersection of at least two visible features with respect to each other;
applying a second video feed of the second player to the machine learning system to detect a second player event;
generating an animation of the second player, the animation having attributes based on motion of the second player from the second player event; and;
automatically providing an enhanced video feed that represents the animation of the second player within the at least one first player event, the animation being represented within a context of the at least one first player event and the attributes of motion of the second player, the context being based on characteristics of at least one spatiotemporal pattern used to detect the at least one first player event.
2. The method of claim 1 , wherein the context is based on characteristics of a plurality of spatiotemporal patterns associated with the at least one first player event.
3. The method of claim 1 , further comprising providing a comparison of a playing style of the first player with a playing style of the second player based on a similarity of at least one spatiotemporal pattern associated with each of the at least one first player event and the second player event.
4. The method of claim 3 , wherein the comparison is based on a similarity of a playing style of the first player, as determined by the machine learning system, with at least one feature of the playing style of the second player.
5. The method of claim 3 , further comprising presenting at least one metric based on the comparison in the enhanced video feed.
6. The method of claim 5 , wherein the at least one metric is spatiotemporally aligned in the enhanced video feed with at least one feature of the at least one video feed on which the similarity is based.
7. The method of claim 1 , wherein the training is further based on a player state transition model that combines transition probabilities of the player state transition model learned empirically from training data, and observed features of movement of the first player.
8. A system comprising:
at least one computer comprising at least one processor and at least one memory, the at least one processor configured to:
process at least one video feed of a first player through at least one spatiotemporal pattern recognition algorithm that is trained through a use of a machine learning system to detect at least one first player event within the at least one video feed, the training based on adjusting an input feature and a weight within the machine learning system, wherein the input is selected from the group consisting of:
relative direction of motion of at least two visible features,
duration of relative motion of visible features with respect to each other,
rate of motion of at least two visible features with respect to each other,
acceleration of motion of at least two visible features with respect to each other, and
projected point of intersection of at least two visible features with respect to each other;
apply a second video feed of a second player to the machine learning system to detect a second player event;
generate an animation of the second player, the animation having attributes based on motion of the second player from the second player event; and
automatically provide an enhanced video feed that represents the animation of the second player within the at least one first player event, the animation being represented within a context of the at least one first player event and the attributes of motion of the second player, the context being based on characteristics of at least one spatiotemporal pattern used to detect the at least one first player event.
9. The system of claim 8 , wherein the context is based on characteristics of a plurality of spatiotemporal patterns associated with the at least one first player event.
10. The system of claim 8 , wherein the at least one processor further provides a comparison of a playing style of the first player with a playing style of the second player based on a similarity of at least one spatiotemporal pattern associated with each of the at least one first player event and the second player event.
11. The system of claim 10 , wherein the comparison is based on a similarity of the playing style of the first player, as determined by the machine learning system, with at least one feature of the playing style of the second player.
12. The system of claim 10 , wherein the at least one processor further presents at least one metric based on the comparison in the enhanced video feed.
13. The system of claim 12 , wherein the at least one metric is spatiotemporally aligned in the enhanced video feed with at least one feature of the at least one video feed on which the similarity is based.