Video loops and management thereof
Systems and methods are configured to identify replacement content within various types of media and to facilitate the use of the identified replacement content in the generation of composited content. The generation of composited content optionally includes Chroma key processes. The content may include video and or multi-dimensional virtual environments. For example, loops of replacement content may be identified within video or time variant virtual environments. These capabilities are optionally facilitated by various artificial intelligence systems, e.g., trained neural networks.
1 . A media system comprising:
storage configured to store replacement content, or to store indexes to replacement content, the replacement content including video, still images, computer-generated virtual environments, mixed-reality environments or time variant virtual environments;
loop logic configured to identify preferred loops within the replacement content;
search logic configured to search the preferred loops based on received search criteria and tags characterizing the preferred loops;
sharing logic configured to distribute at least one of the preferred loops to client devices;
dynamic splicing logic configured to apply a moving frame divider to at least three video frames, the moving frame divider being configured to separate contributions to a splice of the replacement content from the start and the end of one of the preferred loops; and
a microprocessor configured to execute at least the loop logic, search logic, sharing logic or dynamic splicing logic.
2 . The system of claim 1 , wherein the preferred loops are identified based on a quality of an associated loop point.
3 . The system of claim 1 , further comprising video modification logic configured to edit content to create preferred loops.
4 . The system of claim 1 , further comprising tagging logic configured to generate tags characterizing the replacement content, the tags including scenery descriptions, descriptors of elements within the replacement content, search words, or descriptive information configured for searching for the replacement content.
5 . The system claim 4 , wherein the tagging logic is configured to tag multiple preferred loops within a video or time variant virtual environment.
6 . The system of claim 1 , wherein the search logic is configured to identify the preferred loops based on lengths of the preferred loops.
7 . The system of claim 1 , wherein the loop logic is configured to identify the preferred loops by matching an end and a start of a loop.
8 . The system of claim 1 , wherein the loop logic includes a neural network trained to identify preferred loops within replacement content.
9 . The system of claim 1 , wherein the dynamic splicing logic includes a machine learning system trained to select contributions to the at least three frames from the start and the end of the one of the preferred loops, so as to minimize discontinuities within the splice resulting of objects moving between the at least three frames.
10 . A method of providing a preferred loop including replacement content, the method comprising:
receiving a request for the replacement content the request including one or more search terms;
identifying preferred loops that satisfy the received request based on the one or more search terms;
retrieving at least one of the identified preferred loops; and
delivering the at least one of identified preferred loops to a source of the request, wherein the identified preferred loops are associated with a score representing a loop point quality of the preferred loops.
11 . The method of claim 10 , wherein the request includes a minimum preferred loop score.
12 . The method of claim 10 , further comprising modifying one or more of the preferred loops, the modification including dynamic splicing of the preferred loops.
13 . A method of providing a preferred loop including replacement content, the method comprising:
receiving a request for the replacement content the request including one or more search terms;
identifying preferred loops that satisfy the received request based on the one or more search terms;
retrieving at least one of the identified preferred loops; and
delivering the at least one of identified preferred loops to a source of the request, and using one or more of the preferred loops to generate composited content in a Chroma key process.
14 . The method of claim 13 , further comprising modifying imperfect pixels in a monochromatic matte as captured by a camera and changing the imperfect pixels color to a desired color to produce improved composited content including at least one of the preferred loops.
15 . The method of claim 13 , wherein the at least one of the preferred loops includes or is based on a virtual three-dimensional representation of an environment, the representation varying with time.
16 . The method of claim 13 , wherein the replacement content is generated using a virtual environment.
17 . The method of claim 13 , further comprising generating the replacement content using a trained machine learning system.
18 . The method of claim 13 , further comprising applying a moving frame divider to at least three video frames, the moving frame divider being configured to separate contributions to a splice of the replacement content from the start and the end of one of the preferred loops.
19 . The method of claim 13 , wherein the preferred loops are further identified based on a quality of an associated loop point, identified based on a lack of discontinuities at an associated loop point, identified based on positions of objects within identified based on a realistic transition at a loop point.
20 . The method of claim 13 , wherein the preferred loops are associated with tags configured to identify content or characteristics of the preferred loops.
21 . The method of claim 13 wherein the identified preferred loops are associated with a score representing a loop point quality of the preferred loops.
22 . The system of claim 13 , wherein the preferred loops are identified based on a lack of discontinuities at an associated loop point or identified based on positions of objects within the received content.
23 . The system of claim 1 , wherein the preferred loops are identified based on a realistic transition at a loop point.
24 . The system of claim 1 , wherein the search logic is configured to identify the preferred loops based on lighting characteristics of the replacement content.
25 . The system of claim 1 , wherein the search logic is configured to identify the preferred loops based on quality of a loop points within a preferred loop, the quality of the loop points including a smoothness of a transitions between ends of the preferred loops and starts of the preferred loops.
26 . The method of claim 10 , further comprising modifying one or more of the preferred loops, the modification including adjusting brightness and/or color of the preferred loops, changing resolution of the preferred loops, changing file type of the preferred loops, or matching characteristics of the preferred loops to foreground content.
27 . The method of claim 10 , further comprising modifying one or more of the preferred loops, the modification including matching characteristics of the preferred loops to foreground content.
28 . The method of claim 10 , further comprising generating the replacement content using a trained machine learning system.
29 . The method of claim 13 , further comprising modifying one or more of the preferred loops, the modification including dynamic splicing of the preferred loops.
30 . The method of claim 27 , wherein the modification of one or more of the preferred loops is performed in real-time in response to time varying characteristics of foreground content.