The first time a client told me my retouching looked “plastic,” I was mortified. I’d spent three hours on a single portrait, brightening the irises, sharpening every lash, whitening the sclera until it practically glowed. The subject looked like a doll someone had forgotten to finish. That comment was a turning point, and eyes were the culprit every single time.
Eyes are the first thing a viewer’s brain processes in a portrait. Before the skin, before the composition, before the light, the brain goes straight to the eyes and asks: is this person real? When we over-process them, we short-circuit that instinct. The image becomes technically impressive and emotionally inert. Getting eyes right means understanding why they fail in the first place.
Why Eyes Fall Apart in Raw Files
A camera sensor flattens three-dimensional light into a two-dimensional grid of values. The eye, which is a reflective, curved, wet surface sitting inside a shadowed socket, gets compressed into ambiguity. The catchlights lose their crispness. The iris detail, which the brain reads as a signal of vitality, turns muddy. The whites, which are rarely actually white in real life, record as grey. None of this means the photographer did anything wrong. It just means the sensor didn’t capture what the eye perceived.
This is why we retouch. We’re not creating something fake. We’re restoring the perceptual truth that the camera flattened.
The danger is overcorrection. The sclera in a living eye has texture, faint vessels, a slight warm cast near the iris. The iris has depth, not just color. The pupil absorbs light rather than reflecting it. When we ignore these properties and simply crank up brightness and saturation, we produce eyes that read as artificial because they violate the physics of how eyes actually look.
The Layer Stack I Use on Every Portrait
My eye enhancement routine runs inside a single group I call “Eyes - Interstellar” (yes, I name all my Photoshop actions after films, it’s a system that works for me). Inside that group, I work in this order:
1. Sclera correction. I create a Hue/Saturation adjustment layer clipped to a rough selection of the whites. I desaturate the reds by -15 to -25 depending on how bloodshot the eyes are, and I drop the yellows by -10. I never increase brightness here. Instead, I add a Curves layer above it and lift the midtones gently, targeting the range between 100-180 on the input. The result looks like real eyes, not bathroom tiles.
2. Iris depth. I use a Multiply blend mode layer at 20-30% opacity, painted carefully into the outer ring of the iris with a soft brush at 15% flow. This deepens the limbal ring, which is the dark circle at the edge of the iris. Research in visual perception shows that a defined limbal ring reads as a marker of youth and health. We’re not inventing it; we’re restoring what the flat sensor compressed.
3. Iris color and vibrancy. A Vibrance adjustment layer, not Saturation, targeted to the iris. Vibrance protects skin tones while boosting the more muted hues in the iris. I keep this between +15 and +30. Anything higher and the color starts looking painted.
4. Catchlight refinement. On a new layer set to Screen at 40-60% opacity, I use a 10-15px soft brush to gently reinforce the main catchlight. One stroke, maybe two. The goal is a point of light that has a slight falloff, not a hard white circle.
5. Sharpening. I run a High Pass filter at 1.5-2.0 pixels on a merged stamp layer, set to Overlay blend mode, masked to the iris only. This adds micro-contrast that reads as detail without introducing noise.
The whole process takes 15 to 20 minutes on a single portrait once you’ve internalized the steps.
The Mistake I See Most Often
New retouchers almost universally overbrighten the sclera. I understand why: white eyes feel clean, finished, professional. But the human sclera sits in shadow. The upper lid casts a gradient across the top of the eye. The inner corner is almost always a warm pink-grey. When we ignore this and paint the entire white area to a uniform brightness, the eye stops existing in three-dimensional space. It floats.
The fix is simple: mask your sclera adjustments so they affect only the center of the visible white, leaving the shadow areas under the lid untouched. A radial gradient on your mask, feathered at around 50%, handles this automatically.
Matching the Enhancement to the Image’s Mood
This is where I’d push back against the idea of a one-size workflow. Editorial portraits tolerate more visible enhancement than a candid lifestyle image. A high-fashion shot lit by a single dramatic strobe can carry a sharper, more intense iris treatment than a soft natural-light portrait from a family session.
Before I start on eyes, I look at the overall contrast of the image. I ask myself what the photographer was going for. The eye work should feel like it was created in the same room as the rest of the edit. I’ve seen technically flawless eye retouching completely destroy an image because the eyes were too crisp, too saturated, too present for the dreamy, low-contrast mood of the shot.
A useful rule: if the eyes are the first thing you notice when you look at the finished image, you’ve probably gone too far.
The Test That Keeps Me Honest
Before I deliver any portrait, I toggle the visibility of my entire eye group off and on three times in quick succession. I’m looking for a clear difference, but not a shock. If I toggle it off and think “oh, something’s wrong,” the edit is working. If I toggle it on and think “oh, that’s what eyes look like on a poster,” I go back in and walk everything down by 30-40%.
The camera captures light. Our job is to restore life. Those are different things, and the distance between them is exactly as much retouching as the image needs, and not one brush stroke more.
Comments (1)
I tried this on a client project yesterday and the results were way better than expected.
Leave a Comment