This week I want to try something a little bit different on the blog and just see how it goes. Today I’m doing a Photoshop tutorial, or more of a walkthrough, for my shot of my Audiotechnica Headphones. I’ll share my thought process behind a lot of the edits I make and why I make them.
Here’s a before and after GIF
I’ll show you how I got there.
I made a note to take a shot of my light setup before I packed up at the end of the shoot so you can get an idea of where things were placed.
A few things I made note of while I was shooting:
I knew I wanted the final image to be suspended in air without the clamps so I shot a separate image, pretty close to the same angle as the original, to get the top of the headphones without the clamps.
It is always better to shoot images like this to pull elements from instead of trying to clone stamp the clamps out. This way I’m still using real information from the actual headphones instead of trying to make something that isn’t there. Depending on what I’m shooting there may be more than one of these secondary images. I probably could’ve done without this one and just worked from one file, but doing this will make things easier in post.
I also shoot with a ColorChecker Passport whenever possible to create myself a custom camera profile for each individual shoot and set white balance. I also just noticed I was holding this passport upside down but that doesn’t impact it’s functionality at all, I just have to set the crop marks manually in the color checker software. More information on the Colorchecker HERE.
I also shot a blank of my background which is usually a good idea but I found it easier to just make the background from scratch in Photoshop. Having a blank background still gives me a reference image to pull colors from if I feel the need to. We’ll get to that in a few minutes.
So now that I have all the images I’ll be using, I’ll make sure to sync all my changes in Lightroom so that the changes my color checker made to my camera calibration and my custom white balance are the same for all my files. Also any other adjustments like exposure are nice to have synced across all my files as well, but we can also do that in photoshop if necessary.
Now that that’s done I selected all my files (click on first file in the library panel, control click on all the files you want to select) and opened them “As Layers In Photoshop”
Also worth noting that I’m editing in 16-bit ProPhotoRGB to minimize artifacts among other things. This does slow your computer down a bit but it’s better to work with the most information as long as you can and then edit downward on export.
In Photoshop I made a radial gradient using the gradient tool (G) and I sampled the two colors the gradient should use from the blank background I shot earlier.
With the gradient tool still selected you can hold shift and drag from the center and you end up with something like this.
If you don’t create the gradient you want pretty quickly you can make a Curves adjustment layer near the bottom of the layers panel and make an S-curve by clicking and dragging. This gives you some more artifacts in lower bit-depths but more control while you’re editing. If I want to make the gradient bigger or smaller now I can just change the opacity of my Curves layer.
So one of the dead giveaways of things created in photoshop is that they have no noise. Noise is usually something avoided like the plague, the less there is the better, but you do want to make a gradient that looks like it was captured by your camera. Adding noise essentially helps to match the gradient you just created to the pixels created by your camera. Make a new layer (Shift+Cmd+Opt+N) and fill it with 50% Gray (Shift+F5)
Add a small amount of noise for right now using Filter>Noise>Add Noise.
Change the blending mode of the 50% gray layer that you just put noise on to Overlay.
This way, we now have our noise on a layer that is completely independent from our gradient. This way you can duplicate the layer (Cmd+J) and make more noise if you need to, or reduce the opacity if you need to have less. Although lowering the opacity of a noise layer doesn’t look exactly the same as choosing a higher percentage in the noise filter dialogue box. Your choice.
Now there’s multiple ways of doing it but these headphones are perfect for the pen tool. The headphones are made with nice smooth curved lines and crisp edges which is exactly what the pen tool is good for. At the same time, I didn’t shoot on a backdrop that contrasts with the color of the headphones to give me other selection options. Otherwise I could probably use select color range or the magic wand. The pen tool is really the best option, so why not. I’m not going to go into detail on how to use the pen tool, I will point you in the right direction though which happens to be over at Phlearn.com.
So since I used the pen tool for the path around the headphones I went and saved it in the Paths palette by renaming it. This way if I need to make that selection again or change the path, it’s saved.
With my selection still active I created a layer mask which loaded my selection as the white (revealed) part of the layer mask. Above that layer I added the shot I took of my headphones without the clamps on them. I made another layer mask and filled it with black (hide all) and painted white with the brush tool on the top part where I want to use the leather to cover the clamps. It didn’t match up perfectly straight away so I used transform (Cmd+T) to move the layer into place.
The sharpen layer is clipped to the part of the new headphones layer covering the clamps because in order for it to fit nicely it was up-scaled a bit. Up-scaling is also usually not something to do regularly but it’s okay if it’s just a little bit. The sharpen layer is just to keep the texture of the top of the headphones consistent between itself and the layer below. Those layers are grouped together to keep things organized.
So now we have headphones to work with all together in one group. We’re almost done.
I made a new layer to start retouching and clean some things up. Even though I photographed the headphones without the clamps on them, there’s still lines in the leather where the clamps where and they’re a little misshapen but went back to normal after a few days.
At this point to clean this up you can kind of pick your poison of the retouching tools. For tighter edges it’s better to use the clone stamp because it just copies over the sample area. For pretty much anywhere else I prefer using the healing brush which is essentially the same as the clone stamp except it also does a little bit of blending. I like to work on new layers as much as possible and the clone stamp and healing brushes enable me to work on new layers. The patch tool does not.
After a few minutes things are looking like this.
Cool, but the leather itself looks like it should be almost a U-curve but it’s not. The liquify tool is good for fixing this but we need to get all the work we’ve done on the headphones up to now onto a new layer.
I got all the layers that had to do with the headphones together and grouped them (Cmd+G). Then you can duplicate the group by holding option and dragging upwards on the group, right clicking and selecting “Merge Group” and then hiding the old group we worked on.
Now the headphones should be in one piece on a new layer with nothing else. I opened up the liquify tool (Shift+Cmd+X), grabbed the Forward Warp tool, which is the tool you’ll use about 90% of the time in here) and brought the brush size up to push out the top of the headphones.
So we’re just about ready to export at this point but I’m noticing that the left headphone has too much yellow on it. I’m chalking it up to there’s an older bulb in studio strobe that I used that caused it to be more yellow on that part of the image. My second guess would be that a gold reflector is on the strobe instead of silver reflector, but that can’t be because there was no reflector on the strobe, there was a softbox, and no gels. Anyway, so we gotta get rid of it now.
Create a new Curves layer and fill the layer mask with black. Paint white over the yellow part of the headphones, go into the blue channel and push up right in the middle.
So the point of this is to match the color of the left headphone with the right because no other part of the image is as yellow as this one, so we’re changing it. After painting I dialed it in really close by changing the opacity. You can really push the blue channel really far up if you want so you can see what you’re doing and then bring it down later too if you want.
Now, we have output sharpening. I like doing high pass sharpening because it’s easy and simple. Make a new layer and hold Shift+Opt+Cmd+E to Stamp Visible on the layer. This takes a picture of all the layers you have visible and puts them all together on one layer. Go to Filter>Other>High Pass and choose a radius of two or three ish so that you can really only see the fine details that you want to sharpen.
Change the blending mode to Overlay, Soft Light, or Linear Light and change the opacity to your liking and you’re done, my friend. You can also desaturate your sharpen layer if you have lots of colors in your image and are worried about them changing with this technique.
Save your working PSD and then right click your background layer and click “Flatten Image” click OK to discard hidden layers. Hold Shift+Option+Cmd+S to open the Save For Web Dialogue and make sure it looks like this and you’re good.
When you go to close Photoshop now it will ask if you’d like to save the changes you’ve made to your PSD but remember we already saved it. Click NO because you don’t want to save your flattened image in your working file instead of all the layers and adjustments you made. DONE!
Here’s the final image again:
If you’re reading this now, thanks for hanging out so long, maybe I’ll do another soon. Any questions feel free to ask!