Depth map & 3D photo
The Depth Anything AI estimates from a perfectly normal photo how far away every point is. From that you get a moving 3D photo, a depth map for Blender & co. or a portrait effect with a blurred background – like on a phone, just afterwards. Everything happens in your browser, with no upload and no third parties.
Tap the part of the image that should stay sharp. With every new photo the focus starts on the subject in the middle.
Your browser does not support WebGL – the 3D view is not available.
AI model Depth Anything V2 · one-time ≈ 25 MB from mytoolio.org
No photo yet. Drop an image here – the AI estimates how far away every point is and turns it into a moving 3D photo.
How it works
Choose a photo
Drop a photo into the box, pick it or paste it with Ctrl+V. The AI estimates the depth – in a second with a graphics card.
Pick a view
3D photo: move the mouse or your finger, tilt your phone or let it circle, swing or zoom. Depth map: grey or coloured. Blur: tap what should stay sharp.
Save
The 3D photo as a 4-second video (MP4 or WebM) or GIF, the depth map as PNG – in 16 bit, too – or the portrait image at full resolution.
Frequently asked questions
Are my photos uploaded?+
No. Your photo is only read in your browser and analysed there by the AI – it never leaves your device. The AI model (Depth Anything V2) and the runtime come from mytoolio.org itself; there are no requests to other services.
What is a depth map?+
A greyscale image in which every pixel says how close that point is to the camera – bright is near, dark is far. Such maps are used for 3D effects, as displacement in Blender or other 3D software, for depth-of-field in photo editing or as a guide for AI image generators (ControlNet depth). The AI estimates relative depth, not metres.
What is the 16-bit PNG for?+
A normal PNG only has 256 grey levels, which shows as steps on large surfaces. The 16-bit version has 65,536 levels and is therefore better for displacement in 3D software and for smooth gradients. It is scaled up to the photo size (up to 4,096 pixels).
Why does the 3D photo tear at edges?+
From a single photo nobody knows what is hidden behind a person or an object. When moving, the edges are therefore stretched. With less strength or the focus plane on the main subject this is hardly noticeable; the zoom effect is the least sensitive.
How good is the portrait blur?+
Very convincing for clear subjects in front of a calm background. The depth map is fitted to the edges in the photo for this; very fine details such as single hairs or fences may still get a little blurred. It is saved at full resolution, as JPG (PNG stays PNG).
How long does it take?+
With WebGPU (current Chrome or Edge on a computer, many Android devices) the graphics card estimates the depth in about a second. In other browsers the processor does it – about 10 to 20 seconds. The 3D effect and the blur then always run smoothly with WebGL.
Depth estimation with Depth Anything V2 Small (Apache 2.0, weights stored as int8) via ONNX Runtime Web (MIT) – both served by mytoolio.org and run entirely in your browser. The depth is an estimate and not to scale.