Eigenvectors and gradients


Here we show how the diffusion maps, approximating the diffusion distances, look like. The first eigenvector differentiates between the sky and the field, as indicated by the different sign assignment. The second vector differentiates between the red flowers and the green stems and the third between he clouds and the blue sky. The gradients magnitudes according to the diffusion distance is shown for different values of t. As this value increases, the influence of eigenvectors that correspond to smaller eigenvalues become weaker.

Instructions: Roll your mouse over each thumbnail to see a larger version of the image appear below. This makes it possible to easily flip back and forth between the images. Click on a thumbnail for a full resolution version of the image.


Example Image    
Lambda = 0.99 Lambda = 0.96 Lambda = 0.90
1 1
t=16 t=64 t=256