Dynamic Photo HDR

28 Dynamic Photo HDR

Here is the difference between processing (1) Smoother, (2) Sharper on a misaligned image (1:1 pixel

view):

Custom

You can draw a graph of new weighting function. This refines the way how the pixels in different

exposures are weighted.

The horizontal scale represents the whole dynamic range. Normally you would prefer more pixels from

the center (where the image is well exposed) than pixels from the sides (under/over exposed).

Few examples of weighting function:

(1) Hat function that cuts off only the extreme over/under exposed pixels

(2) Over-exposed fix - for example if all exposures have certain over-exposed parts, you may want to

cut the over-exposed pixels much earlier.

(3) Gaussian function favors more center of exposure range.

(4) Pyramid function - linear attenuation of exposures

Unless you closely watch the result pixels, you will not see dramatic changes in the final HDR. It is

more subtle.

Inverted Camera Response