Spiderman visited Zürich!

One of the direct consequences of my Scavenger trip to Orlando (more about that later) is that I got weirdly motivated to start learning “proper” photo editing/processing. And hence, here comes my first attempt at that – and a bit of write-up, both to not forget what I did, and to share with whoever may find it useful.

I started with a picture of Spiderman taken during the workshop:

The picture itself is a Spiderman action figure, held with a piece of 8-gauge wire to some box in the room, and lit with a Lume cube (I’m DEFINITELY getting a couple of these) held in the hand that was not holding the camera. This is already processed a tiny bit in Darktable – levels, (debatable) white balance, that sort of things. I exported in TIFF to not start compressing before I had to, but I’m not sure this was absolutely necessary.

Then I looked into my archives for places where I could put Spiderman, and I ran into this older picture, snagged with my phone a while ago to show snow falling quite late in the year:

Now that I had something that I thought could work, time to start GIMP and work on it! The first step was getting a cleaner image of Spiderman, removing the background. I first cropped the image to only contain the part I was interested in, added a layer mask to my image, and worked with the paintbrush tool (with finer and finer brushes) to remove all the stuff that was Not Spiderman. I left the bit of wire that held the figure on the waist, I figured it was not that visible and could be confused with a belt, or something. In order to keep the layer and the mask (if I wanted to go back to the picture), I created a new layer from the “visible content”, and that’s what I copy in the other image. (I have no idea if that’s the “canonical” way of doing things, but it worked that way for me 😉 ).

Then I pasted and scaled/positioned the figure in the image, doing my best to match the lights in both images.

I did make a mistake there: I realized that my clipping was not great (there was still a lot of stuff around the legs, in particular), but instead of correcting it in the initial image, I started correcting it in the composite. Retrospectively, it doesn’t make much sense; I’ll blame jet lag.

I blurred the background so that it would look better (especially since Spidey’s legs DO have focus blur) – I’m not sure if it was blurred enough, it’s tough to tell. I also re-added a couple of brighter spots in the image to better match the light on Spiderman, which I think kind of helped, but that I might have over done it a bit. I also added a bit light below my pasted Spiderman. I also tried my best to better match the color temperature of both images – I think it’s okay.

Finally, Spiderman cannot be complete without a couple of spider webs – especially since his hands are very much in position! These ones were paths, stroked in grey and edited with a large white brush to make them a little bit less uniform. I tried playing with the gradient tool (which seemed…. made for the task), but failed at it. Then, I decreased the opacity of the “web” layer, and added a bit of blur to it. The spider webs are in my opinion definitely the weakest part of the image, and I kind of believe that making them better would require better drawing skills than mine 😉

Still, for a first attempt at this sort of things, I’m super pleased with the result 🙂