Abstract: In one embodiment, a method includes recognizing an object in an image that is captured by a camera and presented in a region of a screen of the computing device, generating a 3-dimensional mesh representation for the object by recognizing visual components of the object, where the 3-dimensional mesh representation comprises a plurality of polygons, receiving one or more inputs from the user, where the inputs cause color information for at least a part of the region of the screen to be updated, identifying one of the plurality of polygons that corresponds to a first region of the screen, identifying an area of the texture layer that corresponds to the identified polygon, recording the updated color information in the identified area of the texture layer, generating an augmented reality effect associated with the object based on the updated color information recorded in the identified area of the texture layer.