A text prediction system and method for predicting entered text from a captured image is disclosed. An image capture device captures an image with within a field of view for the image capture device where the image has at least one object in the image. The image capture device identifies at least one object in the captured image. A set of words are identified for at least one object in the captured image. This set of words is then provided to a text entry device to be presented to the user as a possible suggestion.