Skip to content

Latest commit

 

History

History
10 lines (8 loc) · 780 Bytes

readme.md

File metadata and controls

10 lines (8 loc) · 780 Bytes
Raw

Image Captioning using Attention Mechanism with BLEU Score evaluation

Implemeted the Encoder-Decoder mechanism of Image Captioning using Attention Mechnism. The model was trained on the Flickr8k dataset.

  • The encoder model uses InceptionV3 as base model to extract features from the images.
  • The attention part links up the extracted features with the captions of the image.
  • Decoder is implemented using GRU.
  • Model training checkpoints can be found here.

The model obtained an average BLEU score of 0.561036 on the validation data of 7000 captions.