Smart closed caption positioning system for video content

A system for generating closed captions via optimal positioning and character specific output styles, the system comprising one or more processors configured to perform a method comprising:

• receiving, by the system, a video input;
• generating, by the system, closed caption data from the video input based at least in part on extracting text data from an audio portion of the video input;
• and
• for each given frame of the video input that has closed caption data associated with the given frame:
• identifying, by the system via facial recognition and audio tone matching, one or more characters who are speaking within the given frame;
• obtaining, by the system, a respective text style for each given character that uniquely identifies the given character from the one or more identified characters;
• generating, by the system, captioning from the closed caption data for each of the one or more identified characters in the respective text style of each of the one or more identified characters;
• extracting, by the system, visual objects from the given frame of the video input;
• determining, by the system, a level of importance of each of the extracted objects of the given frame to video content of the video input to identify one or more important objects of the extracted objects having a level of importance that is higher than an importance threshold, wherein the level of importance of each of the extracted objects is established based at least in part on application of one or more machine learning algorithms;
• comparing visual data of the given frame with visual data of a previous frame to determine a degree of visual similarity;
• and
• upon the degree of visual similarity between the given frame and the previous frame being below a visual similarity threshold, identifying one or more areas of the given frame that do not overlap with the identified one or more characters and the identified one or more important objects, selecting an area of the one or more areas that is positioned at a text insertion location that is close to a text insertion location of the previous frame, inserting, by the system, the generated captioning into the given frame within the selected area.

A non-transitory computer program product for generating closed captions via optimal positioning and character specific output styles, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a system comprising one or more processors to cause the system to perform a method comprising:

• receiving, by the system, a video input;
• generating, by the system, closed caption data from the video input based at least in part on extracting text data from an audio portion of the video input;
• and
• for each given frame of the video input that has closed caption data associated with the given frame:
• identifying, by the system via facial recognition and audio tone matching, one or more characters who are speaking within the given frame;
• obtaining, by the system, a respective text style for each given character that uniquely identifies the given character from the one or more identified characters;
• generating, by the system, captioning from the closed caption data for each of the one or more identified characters in the respective text style of each of the one or more identified characters;
• extracting, by the system, visual objects from the given frame of the video input;
• determining, by the system, a level of importance of each of the extracted objects of the given frame to video content of the video input to identify one or more important objects of the extracted objects having a level of importance that is higher than an importance threshold, wherein the level of importance of each of the extracted objects is established based at least in part on application of one or more machine learning algorithms;
• comparing visual data of the given frame with visual data of a previous frame to determine a degree of visual similarity;
• and
• upon the degree of visual similarity between the given frame and the previous frame being below a visual similarity threshold, identifying one or more areas of the given frame that do not overlap with the identified one or more characters and the identified one or more important objects, selecting an area of the one or more areas that is positioned at a text insertion location that is close to a text insertion location of the previous frame, inserting, by the system, the generated captioning into the given frame within the selected area.