Pareidolia, but underwater: What is under the green paint?

[oshfdk]

This is very similar to what I tried last year, the only difference is instead of single pixels I used a diamond-shaped kernel of 3x3 or 5x5 pixels, concatenating its pixel vectors in both RGB and Lab, and using linear regression. Unfortunately, the result wasn't particularly remarkable. E.g., the same lady:

   

The full gallery: You are not allowed to view links. Register or Login to view.