Intrinsic Image Decompositions

This page lists the decompositions for all 5230 photos, using this algorithm:

Sean Bell, Kavita Bala, Noah Snavely. "Intrinsic Images in the Wild". ACM Transactions on Graphics (SIGGRAPH 2014). http://intrinsic.cs.cornell.edu.

and these parameters:

abs reflectance weight: 0
abs shading gray point: 0.5
abs shading weight: 500.0
chromaticity weight: 0
kmeans intensity scale: 0.5
kmeans n clusters: 20
n iters: 25
pairwise intensity chromaticity: True
pairwise weight: 10⁴
shading blur init method: none
shading blur sigma: 0.1
shading target norm: L2
shading target weight: 20000.0
split clusters: True
theta c: 0.025
theta l: 0.1
theta p: 0.1

View all algorithms.

Image

Reflectance

Shading

WHDR: 55.4% (δ: 0.1), Runtime: 144.5 s, Num. judgements: 31. View human reflectance judgements.

Image

Reflectance

Shading

WHDR: 11.5% (δ: 0.1), Runtime: 302.1 s, Num. judgements: 32. View human reflectance judgements.

Image

Reflectance

Shading

WHDR: 17.3% (δ: 0.1), Runtime: 226.1 s, Num. judgements: 32. View human reflectance judgements.

Image

Reflectance

Shading

WHDR: 21.7% (δ: 0.1), Runtime: 173.0 s, Num. judgements: 32. View human reflectance judgements.

Image

Reflectance

Shading

WHDR: 29.9% (δ: 0.1), Runtime: 418.0 s, Num. judgements: 32. View human reflectance judgements.

Image

Reflectance

Shading

WHDR: 34.4% (δ: 0.1), Runtime: 250.5 s, Num. judgements: 32. View human reflectance judgements.

Image

Reflectance

Shading

WHDR: 40.7% (δ: 0.1), Runtime: 164.9 s, Num. judgements: 32. View human reflectance judgements.

Image

Reflectance

Shading

WHDR: 48.9% (δ: 0.1), Runtime: 231.5 s, Num. judgements: 32. View human reflectance judgements.