Project Topic: Image captioning using RNN

Team Name: Sleepy Heads

Input

Output

Model download

wget https://doc-0k-bc-docs.googleusercontent.com/docs/securesc/ha0ro937gcuc7l7deffksulhg5h7mbp1/ugaptvtuqmpggkod3ofpbng8b5jsbnhf/1555293600000/15570688570768808365/*/1nMn__KLw8WMj_6SDRIH8AFfbaFX78mh7?e=download
Place it in models/model directory and rename it as model.npy

To run the model use : python main.py --phase=test --model_file='./models/model/model.npy' --beam_size=3 Input files hould be kept in test/images

Explaination:

We input an image and in the output we get the image with a caption describing the contents of the image. It works on a CNN for extracting features from the image using inceptionV3 and uses RNN for decoding these features into a sentence. This project is implemented on Tensorflow.