search

fukun07 / neural-image-captioning

Using scene-specific contexts and region-based attention in neural image captioning
MIT License
44 stars 18 forks source link

Generating what the attention model is looking at #2

Closed mhashas closed 7 years ago

mhashas commented 7 years ago

I was wondering if there is a way of visualizing what the attention model is paying attention to and what word is generated, such as the 5th figure in your paper.

fukun07 commented 7 years ago

You may also look at our TPAMI version, which provides more examples of attention visualization (figure 8).

In short, we plotted these figures by setting regions' brightness according to their attention weights.