From CS294-69 Image Manipulation and Computational Photography Fa11

Jump to: navigation, search

Name: Viraj Kulkarni

Assignment: Light Fields/HDR/Tone Mapping

Course: CS294-69, Fall 2011, UC Berkeley

Paper Implemented: Photographic Tone Reproduction for Digital Images. Erik Reinhard, Michael Stark, Peter Shirley and James Ferwerda. SIGGRAPH 2002. (pdf)


I have implemented the above tone mapping paper in MATLAB. I used radiance maps which I found online at this link. The program takes a .hdr image as input and tone maps it according to the method described in the paper. There are some values and parameters that you can vary and these affect how the tone mapped image looks.

The two parameters that I have varied are the key-value and the Lwhite value. The key-value effectively determines the region of the print zone spectrum to which dynamic range gets mapped to. The second parameter than can be varied is the Lwhite value. Lwhite is the smallest luminance value that will be mapped to pure white. All luminance values higher than Lwhite will burn out. These parameters and their effects are demonstrated by the following examples.


Below are the tone-mapped outputs.

Changing the key value tone maps the image to a different region of the print zone spectrum. The effect of changing the key value is illustrated below.

Key-value = 1.8, Lwhite = 1.1

a = 1.8, Lwhite = 1.1

Key-value = 3.0, Lwhite = 1.1

a = 3.0, Lwhite = 1.1

Changing the Lwhite value gives an effect of compressing the range. This value represents the threshold luminance beyond which all values will get mapped to white. The effect of changing this value is illustrated below. Lower the value of Lwhite, more is the dynamic range that gets compressed in the 0-255 luminance values of the output image.

Key-value = 1.8, Lwhite = 0.8

a = 1.8, Lwhite = 0.8

Key-value = 1.8, Lwhite = 1.1

a = 1.8, Lwhite = 1.1

Key-value = 1.8, Lwhite = 2.1

a = 1.8, Lwhite = 2.1

[add comment]
Personal tools