Three-dimensional eyeglasses modeling from two-dimensional images

A method, comprising:

• receiving, at least two two-dimensional (2D) images of a pair of eyeglasses, wherein the pair of eyeglasses is offered for sale by a merchant;
• extracting a first set of contours of the pair of eyeglasses from a first image of the at least two 2D images and a second set of contours of the pair of eyeglasses from a second image of the at least two 2D images, wherein the first set of contours and the second set of contours define a plurality of shape features of the pair of eyeglasses;
• determining, based at least in part, on a meeting place between the first set of contours and the second set of contours, locations for hinges of the pair of eyeglasses using the contours;
• determining, using at least one of the two 2D images, one or more visual features of the pair of eyeglasses;
• generating a three-dimensional (3D) model of the pair of eyeglasses using the first set of contours, the second set of contours, the visual features and the determined location of the hinges, the 3D model including one or more arms extending from the determined location of the hinges;
• automatically altering one or more of a size of the one or more arms and adjusting an angle of the one or more arms;
• and
• communicating the 3D model to the merchant for display.

A non-transitory computer-readable storage medium comprising instructions that, when executed by a processing unit, cause the computer system to perform a method, comprising:

• receiving a selection of at least two two-dimensional (2D) images of a pair of eyeglasses displayed on a computing device;
• extracting a first set of contours of the pair of eyeglasses from a first image of the at least two 2D images and a second set of contours of the pair of eyeglasses from a second image of the at least two 2D images;
• determining locations for the hinges of the pair of eyeglasses using a meeting place between the first set of contours and the second set of contours;
• determining parameters of a nose bridge features of the pair of eyeglasses using the first set of contours and the second set of contours;
• determining parameters of one or more arms for the pair of eyeglasses using one or more of the first set of contours and the second set of contours;
• generating a three-dimensional (3D) model of the pair of eyeglasses using the first set of contours, the second set of contours, the one or more arms and the nose bridge feature;
• and
• providing the generated 3D model of the pair of eyeglasses on a user interface of a display device.