High Dynamic Range

In this assignment, you will combine multiple pictures to create a high dynamic range
images and then use bilateral filter to compress its range.

Useful links
http://www.debevec.org/HDRShop/

http://www.debevec.org/Research/HDR/#viewing
Specifically, this paper describes technique to build HDR maps
   http://www.debevec.org/Research/HDR/debevec-siggraph97.pdf



--------------------------------------------

HDR compression is based on this paper
   http://people.csail.mit.edu/fredo/PUBLI/Siggraph2002/

Here is the high level code for HDR compression

luminance= color2grey(image)
r=R/(luminance), g=G/luminance, B=B/luminance
base=bilateralFilter(luminance)
detail=luminance/base
output=contrastReduce(base) * detail
Routput = r*output, etc.


---------------------------Assignment Description---------------

Dataset1
http://www.cs.huji.ac.il/~danix/hdr/results.html

Dataset2
Take atleast 5 pictures with varying exposure

	-------------------------------------------------
Create HDR image
	Average the five images in Matlab
	Use HDRshop to create a floating point high dynamic range images
	Write Matlab code to (a) create your own high dynamic range image
			(b) implement bilateral filtering and compress the image
Submit
	Due Sept 28th
	All code
	For dataset1: final results (three results: average, HDRshop output, compressed image)
	For dataset2: input images and three results
	Upload as single zip file as FirstnameLastnameProg1.zip  to ftp://ftp.merl.com/incoming/csg242


---------------------------End Assignment Description---------------

When you are creating your own high dynamic range image by fusing 5+ images, you can
use a simple scheme (e.g. take largest pixel value) or implement a weighted
combination as described in 
  http://www.debevec.org/Research/HDR/debevec-siggraph97.pdf

-----------------------------------------------------------------------