A method for virtual shoes fitting includes providing two augmented reality (AR) reference papers, each having a plurality of AR markers and a standard size reference object, one reference paper being placed on a ground and the other AR reference paper being placed directly behind and at right angle to the ground; and providing a mobile device having a camera, a processor, and a machine-readable medium with instructions that, when executed, cause the processor to produce AR markers on the mobile device, and automatically taking photos of a user's foot placed on the AR reference paper on the ground together with the standard size reference object. A system for virtual shoes fitting is also disclosed.