My RawTherapee/Enfuse/Gimp workflow

To show my readers my workflow I decided to blog about it. Today I will show you how this image was created:

1. Basics
This tutorial assumes you are running a linux system with:

• RawTherapee 2.4rc2 – Download
• enfuse 3.2 – Download
• gimp 2.6 – www.gimp.org

The raw file can be found here and is also CC-BY licensed: Download

2. RawTherapee

So lets open RawTherapee and load the raw file. I usually don’t change much there, it’s more a matter of personal choice. I use to sharpen and resize afterwards so we’ll only change brightness in RawTherapee for later use with enfuse.

With “Auto Levels” set in the Exposure tab under Exposure we’ll get a brightness value of 1.22 which means I underexposed the image, but that’s no problem. Save this version of the image as IMG_8886a.jpg, change brightness to 1.5 save the result as IMG_8886b.jpg and do the same for brightness 2.0 and IMG_8886c.jpg. If you’ve done it correctly it should look like this:

This is all I do in RawTherapee for this image, now let’s move on to enfuse

3. Enfuse

Borrowed from the panotools wiki:

Enfuse is a command-line program used to merge different exposures of the same scene to produce an image that looks very much like a tonemapped image (without the halos) but requires no creation of an HDR image. Therefore it is much simpler to use and allows the creation of very large multiple exposure panoramas

I use it to get more details from darker parts, like in this case to brighten up the eyes naturally as I don’t like the look of HDR images.

Open a terminal and change to the directory where the three created images are in, now run following command:

enfuse -o out.jpg IMG_8886*.jpg

This should output in the terminal:

enfuse: Input image "IMG_8886a.jpg" does not have an alpha channel. Assuming all pixels should contribute to the final image.
enfuse: Input image "IMG_8886b.jpg" does not have an alpha channel. Assuming all pixels should contribute to the final image.
enfuse: Input image "IMG_8886c.jpg" does not have an alpha channel. Assuming all pixels should contribute to the final image.
Loading next image: IMG_8886a.jpg
Loading next image: IMG_8886b.jpg
Loading next image: IMG_8886c.jpg
Using 9 blending levels
Generating Gaussian pyramid: g0 g1 g2 g3 g4 g5 g6 g7 g8
Generating Laplacian pyramid: l0 l1 l2 l3 l4 l5 l6 l7 l8
Generating Gaussian pyramid: g0 g1 g2 g3 g4 g5 g6 g7 g8
Generating Gaussian pyramid: g0 g1 g2 g3 g4 g5 g6 g7 g8
Generating Laplacian pyramid: l0 l1 l2 l3 l4 l5 l6 l7 l8
Generating Gaussian pyramid: g0 g1 g2 g3 g4 g5 g6 g7 g8
Generating Gaussian pyramid: g0 g1 g2 g3 g4 g5 g6 g7 g8
Generating Laplacian pyramid: l0 l1 l2 l3 l4 l5 l6 l7 l8
Generating Gaussian pyramid: g0 g1 g2 g3 g4 g5 g6 g7 g8
Collapsing Laplacian pyramid: l8 l7 l6 l5 l4 l3 l2 l1 l0

You should have a “out.jpg” in the folder where the three images are in which should like this:

Let’s open the image in gimp:

4. Gimp
To give the image the final touch I open the levels dialog via right click on the image → Colors → Levels:

I change the input level’s left value to 25 and the input level’s right value to 240:

This yields:

Now I crop the image using the rule of thirds guide and putting the upper right crossing right into her left eye:

If you want to resize and sharpen use the unsharp mask, I usually don’t resize so we’re done here. The resulting image varies a bit from the flickr one, due to me creating this one all anew today and using different values in the levels tool probably:

This work is licensed under a Creative Commons Attribution 3.0 Germany License

If you like my work, consider flattring me. Thank you

About these ads