This class was learned from 131 images of Halloumi. The visualizations below show the maximally activated representations of what makes Halloumi look like Halloumi. 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: 131
- Image sources: Google (72) , Flickr (59)
- Bounding boxes and URLs: halloumi.json