Image Descriptor Generator Using Encoder-Decoder Model

摘要

A visual representation is an image of a situation, object, or person, whereas a description is a description of the picture action. The generation of an image description is critical in many industries, including medicine, advertising, and robotics. This paper considers the well-known deep learning problem of generating a textual description of an image. The proposed model has used InceptionV3and Beam search to run the picture descriptor generator.