In the previous chapters, we have learnt about analyzing sequential data (text) using the Recurrent Neural Network (RNN), and also about analyzing image data using the Convolutional Neural Network (CNN).
In this chapter, we will be learning about using the CNN + RNN combination to solve the following case studies:
- Handwritten-text recognition
- Generating caption from image
Additionally, we will also be learning about a new loss function called Connectionist Temporal Classification (CTC) loss while solving the handwritten-text-recognition problem.
Finally, we will be learning about beam search to come up with plausible alternatives to the generated text, while solving the caption generating from image problem.
