Image to 3D depth effect
An AI model estimates how far away each part of your photo is, then that depth is used to shift pixels into a 3D anaglyph (red/cyan glasses) or side-by-side stereo image.
This is a depth effect, not a 3D scan
The AI model (Depth Anything V2) estimates <em>relative</em> depth — what's nearer or farther within the photo — not real-world measurements. That depth map shifts pixels sideways by an amount proportional to how close they are, which is exactly what creates a stereoscopic illusion when viewed with red/cyan glasses or by cross-eyed viewing. Think of it as a stylized 3D effect rather than a precise 3D reconstruction.
Why does the first conversion take longer?
On first use your browser downloads the depth-estimation model (about 50 MB total) from this site. It's cached afterwards, so later conversions on this page start much faster.
Anaglyph or side-by-side — which should I pick?
Anaglyph needs a pair of red/cyan glasses but works as a normal flat image otherwise. Side-by-side needs no glasses but does need either a VR/stereo viewer or practice viewing cross-eyed — pick whichever you have a way to view.
Is my photo uploaded anywhere?
No — the AI model runs entirely inside your browser (WebAssembly). Your photo never leaves your device.
