Image Captioning ( Use Python Jupyter Notebook )
Prepare one python notebook to build, train and evaluate model (TensorFlow or TensorFlow.Keras library recommended) on the datasets given below.
Image Captioning is the process of generating textual description of an image. It uses both Natural Language Processing and Computer Vision to generate the captions. The dataset will be in the form [image ? captions]. The dataset consists of input images and their corresponding output captions.
Read the pickle file (https://drive.google.com/file/d/1YX_ossqpLuYZxER4b91eFphhKRTf3U6H/view?usp=sharing) and convert the data into the correct format which could be used for ML model.
Pickle file contains the image id and the text associated with the image.
Eg: ‘319847657_2c40e14113.jpg#0tA girl in a purple shirt hold a pillow .
Each image can have multiple captions.
319847657_2c40e14113.jpg -> image name
#0 -> Caption ID
t -> separator between Image name and Image Caption
A girl in a purple shirt hold a pillow . -> Image Caption
Corresponding image wrt image name can be found in the image dataset folder.
Image dataset Folder : https://drive.google.com/file/d/1AZ213vTwLTzRLqbuh-zpv8Jkl_-N2_2m/view?usp=sharing
Plot at least two samples and their captions (use matplotlib/seaborn/any other library).
Bring the train and test data in the required format.
Use Pretrained Resnet-50 model trained on ImageNet dataset (available publicly on google) for image feature extraction.
Create 5 layered LSTM layer model and other relevant layers for image caption generation.
Add L1 regularization to all the LSTM layers.
Add one layer of dropout at the appropriate position and give reasons.
Choose the appropriate activation function for all the layers.
Print the model summary.
Compile the model with the appropriate loss function.
Use an appropriate optimizer. Give reasons for the choice of learning rate and its value.
Train the model for an appropriate number of epochs. Print the train and validation loss for each epoch. Use the appropriate batch size.
Plot the loss and accuracy history graphs for both train and validation set. Print the total time taken for training.
Take a random image from google and generate caption for that image.