README~

1. Run reLight.py with python.
2. tk window should pop up. Default settings should run example data sets.

Crispin Bernier (chb2ab@virginia.edu)

This started as a group project in Jason Lawrence's Computer Vision class (University of Virginia) with group members Jinlong (Frank) Feng, Jack Doerner, Haoran Hou, and I.

This is an implementation of the paper:
Light Waving: Estimating Light Positions From Photographs Alone

Holger Winnemöller	
 

Ankit Mohan

 	
Jack Tumblin

 	
Bruce Gooch


which can be found here: http://web.media.mit.edu/~ankit/lightwaving/


Notes
- Requires scipy
- For doing MDS on a CSV file each row should be a sample and each collumn should be a dimension.
- The slowest part of this is doing floyds algorithm on N, where N is the number of samples.
- The test sets were made in Blender.
- These are descriptions for the buttons along the top of the final window.

w/Neighbors: Shows the neighbor relations between data points at a particular step. Neighbors are connected with a line and data points have approximately k neighbors with k being a user input. Note that w/Neighbors generally takes a much longer time for pyplot to display and is slower to manipulate then the other visualizations.

Raw 3D Embedding: Embedding of the data into 3 dimensions using classic multi-dimensional scaling. This is done after step 3.

Fit 3D Embedding: The Raw 3D embedding is then fit to a sphere using least squares regression and the data is projected onto the sphere. This is also shown in step 4 where the user selects the point to unwrap from.

Unrolled Embedding: This is also displayed on the blue panel in step 7. After projecting the data onto a sphere it is unwrapped into 2 dimensions.

save: Saves a csv file in the same directory with the unrolled embedding.

flip: Flips the points on the control panel vertically

Thank you!! :)