Most images do not have a description, but the human can largely understand them without their detailed captions, but machine needs to understand some form of description of image. In Image Captioning, textual description of an image is generated. These captions could be used for various purposes like automatic image indexing. Image indexing is an important of Content-Based Image Retrieval (CBIR) and hence, it can be applied to many areas, including biomedicine, commerce, the military, education, digital libraries, and web searching. We have divided the task into two parts- one is image based model which is used to extract the content of the image for that purpose we have used CNN model and other a language model which is used to translate the feature in sentences for that purpose we have used RNN(LSTM).
Keywords : CNN, RNN, LSTM, TBIR, CBIR, BLEU.