To use all the features of this site you must be logged in. If you don't have an account you can sign up right now.
The photo rating are based on Google Neural Image Assessment (NIMA) for judging photos using Artificial Intelligence.
Evaluation of picture quality and feel has been a long-standing issue in photo handling. While specialized quality appraisal manages to measure pixel-level debasements, for example, commotion, obscure, pressure antiques, and so forth., tasteful evaluation catches semantic level attributes related to feelings and excellence in pictures. Hossein Talebi, Google Software Engineer said, “Our proposed network can be used to not only score images reliably and with high correlation to human perception, but also it is useful for a variety of labor-intensive and subjective tasks such as intelligent photo editing, optimizing visual quality for increased user engagement, or minimizing perceived visual errors in an imaging pipeline.”
The AI photo tagging are annotated with multiple tags, to enhance the quality of visual representation of the trained CNN model. It is based on a large-scale multi-label image database with 18M images and 11K categories, called Tencent ML-Images. The AI rating assessment and photo tagging will run in 5 mins interval due to compute performance consideration.