Most ways of doing this ask for something you do not have. A chroma key wants a green screen or a plain wall. After Effects wants you to duplicate the clip and roto the subject by hand. Neither is needed.
Try it free in your browser →No sign-up. The clip never leaves your device.
Chroma key works by removing one colour. That is why the tutorials tell you to stand in front of a plain background: the editor is not finding you, it is deleting everything green. Change the background and the effect falls apart.
A segmentation model finds the person instead. It looks at each frame and works out which pixels are the subject, whatever is behind them — a street, a kitchen, a crowd. That is what makes the effect work on a clip you have already filmed, on a phone, with no preparation at all.
Yes. Behindr finds the subject with an AI segmentation model rather than by colour, so it works on a clip filmed anywhere. A green screen is only needed by editors that do the effect with a chroma key.
Yes. The subject is found again on each frame, so the text stays behind them as they move through the shot rather than being masked once and left.
No. That is the After Effects approach, where you duplicate the clip and isolate the foreground by hand. Here it is automatic.
No. The model runs in your browser on your own device, so the clip never leaves it. That also means there is no upload wait and no queue.
Short clips work best. Every frame is processed on your own machine, so a few seconds is quick and a long clip will be slow, particularly on a phone.
The same thing works on a still picture, and that page covers the detail: how to put text behind a person in a photo.
Open Behindr →