Auto Annotation

This is when an AI System automatically assigning metadata to a digital image in the form of captioning or keywords. This approach is a multi-class picture classification with a wide number of levels – as large as the vocabulary scale. Machine learning approaches often use image processing in the form of derived feature vectors and training annotation terms to try to automatically add annotations to new photos. The benefits of automated image annotation over content-based image retrieval (CBIR) include the ability for the user to specify queries more naturally. Images are searched concepts such as color and texture.