We click a button and a background disappears. It's almost too easy to appreciate. But the ability to isolate a subject from its surroundings is something that designers and photographers have been wrestling with for over a century. The journey from physical cutouts to neural networks is an fascinating one.

The scissors era

Before computers, there were X-Acto knives. Graphic artists working on print layouts would physically cut out photographs using sharp blades and alignment grids. The cutouts were then pasted onto a new background and re-photographed. This was slow, error-prone, and left visible edges that had to be manually airbrushed.

Darkroom magic

Photographers found a way to cheat in the dark. High-contrast litho film could be used to create masks that blocked light from specific areas of the print. By carefully combining multiple exposures, they could replace a busy background with pure white. The process was messy, chemical-heavy, and required a trained hand. But it worked.

The digital revolution

When Photoshop 1.0 launched in 1990, the lasso and magic wand tools were introduced. Suddenly, you could select pixels on the screen. The pen tool brought vector-based clipping paths into mainstream design. But these methods still required patience and precision. It could take an hour to trace a complex outline correctly, and it was easy to make mistakes.

The plugin years

By the early 2000s, dedicated plugins like Fluid Mask, Vertus, and Topaz ReMask started using edge-detection algorithms to speed things up. You would paint a rough outline in blue and red, and the software would estimate the boundary. It was a leap, but it was still a long way from a one-click solution.

Enter the AI era

Around 2018, deep learning models like U-Net began to emerge. Trained on millions of labeled images, these models could predict with startling accuracy which pixels belonged to the foreground. The first commercial tool to really nail it was Remove.bg, but now the technology is everywhere – including running in your browser with libraries like @imgly/background-removal.

Today, background removal is no longer a specialized skill. It is a utility, like cropping or resizing. The history of the craft is a reminder that technology doesn’t just speed up tasks, it makes them accessible to everyone.