This class was learned from 158 images of Wine. The visualizations below show the maximally activated representations of what makes Wine look like Wine. Each image begins with a randomly-seeded noise image, which leads to a slightly different class representation. Below are 12 activations created from 12 random noise images.
- Images used for training: 158
- Image sources: Google (76) , Flickr (70)
- Bounding boxes and URLs: wine.json