The Pixel Revolution: How ETH Zurich's New Tech Could Transform Camera Design by 2030

This is one of those times. Research coming out of ETH Zurich, recently published in the prestigious journal Nature, details a completely new type of pixel. It’s a pixel that can both record light and display it. At the same time.

The Short Answer: Researchers at ETH Zurich have created a new pixel, called a Fourier pixel, that can both capture and emit light. This breakthrough could lead to ‘camera-displays’ where your screen is also your sensor, fundamentally changing device design and enabling instant, on-pixel processing within the next decade.

What Exactly is a ‘Bidirectional Pixel’?

For my entire 15+ year career, from the print shop floor to studio photography, a pixel has always been a one-way street. On your camera sensor, pixels are tiny buckets that catch light to form an image. On your monitor, like the Apple Pro Display XDR I design on, pixels are tiny light bulbs that push light out to create an image. One is an eye, the other is a mouth. They don’t do both.

Not anymore.

The team at ETH Zurich, led by Professor David Norris, has created what they call a “Fourier pixel” or “bidirectional pixel.” It’s a single unit that acts as both a camera and a display. It can see the light coming in and project a different beam of light back out, all in one go.

And it does more than just measure brightness. This thing can analyze the very properties of the light wave itself—its phase and its polarization. This is a level of data capture that is completely foreign to conventional consumer cameras.

The Science, Without the Headache

So how does it work? It’s not magic, it’s physics. The core idea is based on light wave interference. When light hits a surface, the waves scatter and overlap, creating complex patterns.

The researchers created a surface sculpted with incredible precision—down to a few nanometers. When incoming light hits this pixel, it’s converted into a special kind of wave that travels across the chip’s surface. This wave is then scattered back out from a different point on the pixel. By controlling the exact shape of that sculpted surface, they can control exactly how the outgoing light waves interfere with each other to form an image.

The math behind figuring out the right surface pattern is called Fourier analysis, which is where the pixel gets its name. This approach is apparently so effective it simplifies the design, bypassing the need for heavy computational modeling.

Why This Actually Matters for Working Creatives

Okay, the science is clever. But what could this mean for those of us in the field, shooting with our Nikons and Godox strobes? Let’s speculate out to 2030.

First, the most obvious application is the “camera-display.” Imagine a smartphone where the entire screen is the camera sensor. No more notches, no more hole-punches. The same surface you watch videos on is the surface that takes the picture. This would be a massive leap in industrial design.

But the real potential is far deeper and, for me, more exciting. The researchers have shown that a Fourier pixel can measure incoming light and emit a corresponding beam back out *in real-time, without a separate computer*. The pixel itself does the processing. Think about that for a second. The lag between seeing, processing, and acting could vanish.

Imagine a camera that doesn’t just have eye-tracking autofocus. Imagine a camera where the sensor itself analyzes the glint in a subject’s eye and simultaneously projects a microscopic beam of fill light to perfectly manage the catchlight, all before the data even thinks about hitting a processor. It’s the difference between a tool that captures light and a tool that has a conversation with it.

Let’s Be Realistic: The View from the Studio Floor

Now for the dose of reality. I’m a working photographer. I care about what survives a shoot in the rain, not what looks good in a lab. This technology is currently a single prototype pixel.

My Nikon Z6 III has 24.5 million pixels. Scaling this tech up from one pixel to millions is a monumental engineering and manufacturing challenge. The nanometer-scale precision required sounds insanely expensive. My years in the print industry taught me one thing: a brilliant design that can’t be produced affordably and reliably at scale is just a nice idea.

So, I don’t expect to see this in a pro camera body anytime soon. The first applications will likely be in highly specialized fields—scientific imaging, autonomous vehicles, maybe augmented reality glasses.

But if they solve the scaling problem? By 2030, we could see devices that look and function nothing like today’s cameras. Maybe a camera becomes a simple, thin sheet of ‘smart glass.’ The distinction between a lens, a sensor, and a viewfinder could blur into a single, intelligent surface. The craft of photography—understanding light, angle, and composition—will always be king. But a tool that can react to light as fast as light itself? That’s not just an incremental update. That’s a new chapter.

The Bottom Line

• It’s a New Ingredient, Not a New Camera: This is a fundamental building block. For now, it’s a scientific breakthrough, not a product. Its real impact will depend on the engineers and designers who figure out how to use it.
• Don’t Sell Your Gear Yet: For working pros, this is a ‘watch and wait’ technology. The path from a single lab pixel to a reliable 45MP sensor in a weather-sealed body is very, very long.
• The Real Revolution is Instant Processing: Merging the camera and display is cool. But eliminating the processing lag between seeing and reacting—by having the pixel itself ‘think’—is the truly profound shift this technology promises.

Frequently Asked Questions

Will this Fourier pixel technology make my current camera obsolete?

Not for a long, long time. This is foundational research and is likely a decade or more away from appearing in any consumer or professional camera you’d actually buy.

Could this technology be used for things other than photography?

Absolutely. The researchers suggest it could be powerful for things like advanced medical imaging, LIDAR sensors for self-driving cars, and next-generation AR/VR glasses.

What is the biggest challenge to making this a real product?

Scaling. Going from a single, perfect pixel in a controlled lab environment to manufacturing millions of them reliably and affordably on a large sensor is the primary obstacle.