So what if we use photos of buildings? does that transition better that faces? For this example I'm using a photograph of the office I work at.

​

For my first example I typed the keywords 'Derelict Building' 'Vegetation' and used a paint-block art style. 

​

I generated this 8 times, and then turn the 8 images into a animated gif, overall the outcome is quite striking and I'm shocked at how well the this turned out.