← Back to Data Science

All Topics

Advertisement

Learn/Data Science/Deep Learning

Image Captioning

Topic: Captioning

Advertisement

Describe Images with Text

Generate image descriptions.

Architecture

Encoder-decoder. CNN for image. RNN/LM for text. Attention.

Show and Tell

NIC model. CNN + LSTM.

Show, Attend and Tell

Attention on image regions.

Key Takeaways

  1. CNN encoder + LSTM decoder
  2. Attention helps
  3. Beam search for better captions

Advertisement

Advertisement

Need More Practice?

Get personalized data science help from ChatWhole's AI-powered platform.

Get Expert Help →