Open your browser camera, point it at your desk or floor, and use gesture-based summoning to bring a character into your actual room on screen — no headset or 3D modeling required.
AR character apps usually ask for one of two things: a compatible headset, or a pre-built 3D model with a tracked marker to anchor it in space. Neither is realistic if you just want a character to show up in your room for a clip or a stream moment. There's a browser-based path that skips both requirements — it works through your regular camera, live, and the character responds to your gestures instead of playing a fixed animation.
This guide walks through exactly how to put a virtual character in your room using nothing but a browser and a camera, plus the setup and framing choices that make the shot look believable.
The whole appeal of this method is how little it asks for. To put a virtual character in your real room, you need three things: a device with a camera (a phone, laptop, or a webcam on a desktop), a modern browser, and a bit of open surface to point at. That's it — no headset, no marker card taped to your desk, no 3D artist, no GPU on your side.
Everything heavy happens in the cloud over a live WebRTC connection. Your camera feed streams up frame-by-frame, the Xmax X2.0 model places and animates the character, and the result renders straight back into a standard video element on your screen in real time. Because there's no download or install step, the same flow works whether you're on a gaming PC or a mid-range phone. This is what makes an AR character with no headset practical: the compute isn't sitting on your hardware.
It's worth being clear about what this is and isn't. You're getting a convincing AR-style result on screen — a character composited into your live camera view — not a depth-mapped 3D reconstruction of your room and not a native mixed-reality experience through glasses. For a clip, a stream overlay, or a social post, the on-screen version is exactly what you want.
You don't need a studio, but a few generic setup choices make the difference between a clip that reads clearly and one that looks muddy.
Getting the character on screen is step one — the fun is in what you stage around it. A few directions worth trying:
Because summoning runs on the same real-time engine as every other mode, you can pivot mid-session — swap the character, change surfaces, or switch to a completely different effect without any render wait.
LiveGen's Summon mode is built exactly for this: it places a virtual character into your real camera feed and keeps it aware of your gestures and position, live, without any 3D modeling, rigging, or headset on your end. Because it runs the same real-time engine as every other LiveGen mode, there's no render wait between summoning the character and seeing it respond — you see the reaction the moment you move. If you'd rather animate a still image than summon a character, Bring to Life gives you a static picture you can direct instead. For more ideas on what to do once a character is in your space, see AI Companion in Your Room.
No. Summon runs through your regular camera and browser — there's no headset or dedicated AR hardware involved.
It tracks your gestures and position continuously through the camera, so its reactions respond to what you actually do rather than repeating a fixed clip.
Yes. Summon works in a mobile browser as well as desktop, using your phone's camera.
No. Summon works through 2D live video — there's no rigging or modeling step. Use a preset character or upload a reference image instead.
Not currently. LiveGen runs in a web browser today, so it's AR-style through your camera and screen — not a native app or AR glasses experience.
It's free to start with free credits, and billing works at 1 credit per second of generation. Free exports are watermarked; paid tiers remove the watermark and add HD. Verify current pricing on the site.
Yes. Upload a reference image of the character you want and Summon uses it as the source, so you can bring an original design or a specific look into your room rather than a stock preset.
It composites into your live camera view so it reads as present in your space on screen, but it's a 2D real-time composite rather than a true 3D object anchored with depth. Even, front-facing light and a clear surface make it look the most grounded.
Open your camera and become anyone — free to start, no sign-up for your first try.
Start generating free