The ML models that have been created for Object Analysis regards the analytics extraction of the inanimate objects of the scene.
These information could be used for the analysis of any object, text or draft which is present on a specific scene.
Possible applications are (in combination with the other components of MuseBox):
Object Detection
It detects the objects in a scene up to 32×32 pixel
Object Recognition
Given a object bounding box, it determines what kind of object is
Text Detection
It detects where is text in a scene
Text Transcription ( OCR )
Given a text bounding box, it transcribes the content
Logo Detection
It detects the logos in a scene up to 64×64 pixel
Logo Recognition
Given a logo bounding box, it determines the brand name