Image-Captioning-Using-CNN-Encoder-and-RNN-Decoder-Computer-Vision

Image Captioning was a project which was developed to provide description of images given to the model. Model Highlights: COCO dataset from Microsoft was used as training dataset which contains around 80 categorical images. CNN (Resnet Architecture) was used as encoder to generate feature vector. Lexical processing was performed using NLTK library RNN decoder was performed using LSTM cells