Character recognition

A method comprising:

• receiving, by a processor, an object detection requirement pertaining to a video clip;
• identifying, by the processor, a visual media feature map from visual media data to process the object detection requirement, the visual media data being obtained from a plurality of frames associated with the video clip;
• implementing, by the processor, an artificial intelligence component to:
• augment the visual media feature map by retrieving ancillary data associated with the video clip from a plurality of sources;
• segment the visual media feature map into a plurality of regions;
• and
• identify a plurality of image proposals from each of the plurality of regions;
• implementing, by the processor, a first cognitive learning operation to:
• allocate a human face identity for a human face present in the plurality of image proposals;
• allocate an object name for an object present in the plurality of image proposals;
• determine a face identity model for the human face present in the plurality of image proposals, the face identity model to include the human face identity and the object name pertinent to the human face identity;
• and
• associate the face identity model for the human face with the pertinent object name and the visual media data for generating a tagged face identity model;
• implementing, by the processor, a second cognitive learning operation to:
• determine whether the tagged face identity model for the human face corroborates with the visual media feature map across the plurality of frames;
• and
• assemble a first frame from the plurality of frames of the video clip with an appurtenant tagged face identity model;
• and
• generating, by the processor, an object detection result corresponding to the object detection requirement, the object detection result comprising the video clip including the application of the appurtenant tagged face identity model across the plurality of frames.

A non-transitory computer readable medium including machine readable instructions that are executable by a processor to:

• receive an object detection requirement pertaining to a video clip;
• identify a visual media feature map from visual media data to process the object detection requirement, the visual media data being obtained from a plurality of frames associated with the video clip and;
• implement an artificial intelligence component to:
• augment the visual media feature map by retrieving ancillary data associated with the video clip from a plurality of sources;
• segment the visual media feature map into a plurality of regions;
• and
• identify a plurality of image proposals from each of the plurality of regions;
• implement a first cognitive learning operation to:
• allocate a human face identity for a human face present in the plurality of image proposals;
• allocate an object name for an object present in the plurality of image proposals;
• determine a face identity model for the human face present in the plurality of image proposals, the face identity model to include the human face identity and the object name pertinent to the human face identity;
• and
• associate the face identity model for the human face with the pertinent object name and the visual media data for generating a tagged face identity model;
• implement a second cognitive learning operation to:
• determine whether the tagged face identity model for the human face corroborates with the visual media feature map across the plurality of frames;
• and
• assemble a first frame from the plurality of frames of the video clip with an appurtenant tagged face identity model;
• and
• generate an object detection result corresponding to the object detection requirement, the object detection result comprising the video clip including the application of the appurtenant tagged face identity model across the plurality of frames.