A Detailed Breakdown Of Shooting A Gokou Ruri 1/8 by Griffon Enterprises Figurine

Gokou Ruri "Nekomimi version" 1/8 scale

This is the first photo I’ve taken to try out the “tinfoil / aluminum foil background” method. Basically, you take a bunch of tinfoil, screw it up into a ball, then carefully unfold it to create a crinkled background. Hang it a good distance behind your subject, then point a flash at it. The crinkles in the foil will create a series of bright reflective points, which when thrown out of focus create a stunning sparkly background.

Background foil

First, let’s take a look at the background on its own. I’ve attached two rows of foil to my wall with sticky tape. I’ve got two flashes pointed at it, the one on the left with an orange gel, and the one on the right with a pink gel. Both flashes are remotely triggered by Yongnuo YN-622C wireless flash triggers. In my final shot, these flashes are firing at about 1/32 to 1/64 power. Not much power is needed because the foil is so highly reflective, in fact the flash on the right is set to its minimum power.

gokou-ruri-i-5NQStKg-XL

Picking an aperture

The choice of shooting aperture is critical for this background, because changing it can cause the appearance to vary from flat and creamy smooth, to individual points of light. I also have to pick an aperture that I have enough light to support, and that will provide enough sharpness for the figure. Here are my options, spread 1 stop apart:

In the end, I chose f/6.3, partway between two of the levels shown above. But if you look closely at that photo, you’ll notice that this doesn’t give enough depth of field to keep the figure sharp from front to back:

gokou-ruri-i-tsv5TrK-X2

Closeup of an f/6.3 frame

In order to get the depth of field I want, I need to take multiple photos at slightly different focus depths, and merge them together (this is called “Focus Stacking”). More details on this later. Now let’s look at the scene overview.

This overview shot was taken from above the position of the tripod, you can just see the tip of the lens hood entering the frame at the bottom.

gokou-ruri-i-JRWjjWQ-X2

Surface

The surface is a glass table, which looks like a mirror from the shallow shooting angle in the final shot (the lens will be about 20cm above the table, putting it at the figure’s eye-line.) If you look closely, you’ll see that I’ve added a sheet of black card underneath the glass to prevent the chairs underneath from showing up in the photo.

Main light

The main light is the cardboard box on the left. At the rear end of the box is an ungelled flash, pointing at white paper towels that I’ve taped over the opening. I’ve used the box’s open flaps to block its light from hitting the background, which would reduce the saturation of the background colour. It’s firing at about 1/8 power.

Rear light

At camera-right is a flash covered with a blue gel, with a mini “speedlite softbox” attached, firing at about 1/16 power. This provides the blue highlight on the right side of the figure.

Laptop

Dust on the figure is very hard to see on the camera’s display, so I take my first few shots tethered to the laptop with Lightroom’s live capture session feature. By zooming in and panning around the image, I can spot where the dust particles are, and then brush them off with a physical lens brush before I start the real shooting.

A closer look at the subject lighting

gokou-ruri-i-xKwfjLM-X2 gokou-ruri-i-VSCdNk6-X2

Shooting

I need to take a series of photos at slightly different focus depths, then merge the sharp parts together, in order to get the figure sharp and detailed from front to back. To take the photos for me I installed the Magic Lantern custom firmware onto my Canon 5D III. This firmware adds a bunch of neat new features, but the one that I’m using is the focus stacking feature.

I use live view to focus on the nearest part of the subject, then use the “rack focus” menu option to move the focus point to the furthest part of the subject. Then I have the Focus Stacking feature copy that Rack Focus range, and sit back and watch it run, capturing 28 photos:

Farthest point, photo 28/28

Farthest point, photo 28/28

Focused on eyes, photo 10/28

Focused on eyes, photo 10/28

Nearest point, photo 1/28

Nearest point, photo 1/28

If your camera doesn’t support this firmware, you can focus manually between shots instead. Or you can connect your camera to your computer, and use your camera’s tethering software (or specialized software like Helicon Remote) to automatically step your camera’s autofocus and take the photos for you.

I use Lightroom to tweak the contrast of the raw photos, export them as 16-bit TIFFs, then open them up in Helicon Focus.

Helicon Focus automatically picks the sharpest parts of each frame and combines them together in order to produce a sharp result. After a bit of playing around with the parameters, I settle on Method B with a Radius of 8 and Smoothing of 4:

Helicon Focus The first frame of the stack is shown on the left, with the final merge result on the right.

Helicon Focus
The first frame of the stack is shown on the left, with the final merge result on the right.

The resulting merged photo and depth map are shown below.

Notice how the variations in the depth map in the background area correspond to ugly artifacts in the background. If you go take a look at the raw frames I shot earlier, you’ll note that the size of the out-of-focus highlights in the background actually changes as the focus changes (they get smaller as you focus further away). Helicon Focus interprets this as detail appearing, and tries to sharpen the background by combining many shots, which I don’t want to happen.

Initial merge result

Initial merge result

Initial depth map

Initial depth map

I use Helicon Focus’ superb built-in retouching tool to paint the first frame in the stack (which, being focused closest to the camera, has the smoothest background) onto the messy original background.

I also make a few other little tweaks (e.g. to clean up the reflections on the base), ending up with the corrected depth map and final result below!

Final depth map after Helicon Focus retouching

Final depth map after Helicon Focus retouching

Gokou Ruri "Nekomimi version" 1/8 scale

Gokou Ruri “Nekomimi version” 1/8 scale

I hope you learned something useful from this behind-the-scenes detail! If you have any questions about it, feel free to leave a comment and get in touch with me!

About The Author

Nicholas Sherlock is a software engineer and a photographer enthusiast based in Dunedin, New Zealand. You can see more of his work in his site. This post was originally published here.

  • http://twitter.com/Potrillo_nin Potrillo

    oye @Bokeeeh si no sigues a esta gente hazlo ya!

  • Alper

    Amazing tutorial. (btw- I would go with f/10)

  • Cedric

    This sounds like a huge amount of digital manipulation. Also taking 28 pictures means triggering the flashes 28 times, not even to set the light.
    What about decreasing the aperture and taking the aluminium foil further from subject ? Or if the space doesn’t allow this, it would eventually be possible to add a diffusion material between the alu foil and the subject to keep a smooth background.

    • Nicholas Sherlock

      The flashes are not particularly powerful (YN-468 IIs), and they’re not firing more than about 1/8 power. 28 shots is hardly anything from these flashes.

      The distance to the foil can’t increased much, the camera is on one side of the room and the foil on the other. Shooting the figure at f/20, it’s clear that the detail in the in-focus areas has been decreased substantially by diffraction compared to f/6.5 (I would estimate a drop of about 40-50% in pixel-level detail), and the tail of the figure is still not within the DOF. It could be done for small web-sized images, but not if I really want to maximise the detail my camera can produce.

  • Paganator

    I really like the wall of bokeh method. Taking two shots with different gels on the background flash and combining them in Photoshop can also give some interesting results, like in this shot I took of a watch: https://www.flickr.com/photos/pagarneau/14268429615/

  • Amaryllis

    Just a question like that, couldn’t you have used a smaller aperture to get the entire figurine in focus, then just one other shot with a larger aperture to get the bokeh, before combining both of them?