Machine learning for recognizing and interpreting embedded information card content

A system for generating a classification model for extracting metadata from a video stream, the system comprising:

• a non-transitory storage medium storing computer program instructions;
• and
• at least one processor configured to execute the computer program instructions to perform operations comprising:
• receiving at least a portion of a video stream;
• identifying one or more card images embedded in one or more video frames of the portion of the video stream;
• processing the one or more card images to extract a plurality of character images, each character image comprising a character;
• generating a training set of character vectors in a vector space based on the extracted plurality of character images;
• reducing dimensionality of the vector space by selecting a subset of the character vectors in the training set;
• and
• generating a classification model to extract metadata from video streams by training the classification model using at least one of the subset of the character vectors.

A non-transitory storage medium storing computer program instructions that when executed by at least one processor perform operations comprising:

• receiving at least a portion of a video stream;
• identifying one or more card images embedded in one or more video frames of the portion of the video stream;
• processing the one or more card images to extract a plurality of character images, each character image comprising a character;
• generating a training set of character vectors in a vector space based on the extracted plurality of character images;
• reducing dimensionality of the vector space by selecting a subset of the character vectors in the training set;
• and
• generating a classification model to extract metadata from video streams by training the classification model using at least one of the subset of the character vectors.