公开(公告)号：US20240073397A1

公开(公告)日：2024-02-29

申请号：US18504355

申请日：2023-11-08

Applicant: Snap Inc.

Inventor： Sagi KATZ ,  Guy MELAMED

IPC: H04N13/183 ,  G06F3/01 ,  G06F3/04815 ,  G06F3/04883 ,  H04N13/128 ,  H04N13/156 ,  H04N13/332 ,  H04N13/371 ,  H04N13/38

CPC classification number: H04N13/183 ,  G06F3/012 ,  G06F3/04815 ,  G06F3/04883 ,  H04N13/128 ,  H04N13/156 ,  H04N13/332 ,  H04N13/371 ,  H04N13/38

Abstract: A wallpaper system presents a first wallpaper image of a wallpaper video to a user and receives, via a user input device, one or both of: (i) a spatial user input selection, and (ii) a time user input selection from the user to apply to the wallpaper video. In response to detecting one or both of: (i) the spatial user input selection, and (ii) the time user input selection, the wallpaper system determines one or both of: (i) a respective spatial movement parameter within a wallpaper video associated with the spatial user input selection, and (ii) a respective time coordinate within the wallpaper video associated with the time user input selection. Wallpaper system presents, via the image display, a second wallpaper image associated with one or both of: (i) the respective spatial movement parameter, and (ii) the respective time coordinate.

公开(公告)号：US20220230284A1

公开(公告)日：2022-07-21

申请号：US17715732

申请日：2022-04-07

Applicant: Snap Inc.

Inventor： Eyal ZAK ,  Guy MELAMED

IPC: G06T5/00 ,  G06T7/13 ,  G06V40/16 ,  G06K9/00 ,  G06T3/40 ,  G06T11/00

Abstract: Systems, methods, and non-transitory computer readable mediums including processes to correct for distortion in images introduced by eyewear (i.e., where the facial region surrounding the eye has a boundary that doesn't match the boundary of an uncovered facial region). The correction includes segmenting images to detect eyewear covered facial regions and facial regions not covered by the eyewear and altering the covered facial regions to match the covered facial boundary to the uncovered facial boundary. Alterations include processing using a machine learning model, applying anti-refraction algorithms, scaling the covered facial region to match boundaries of the uncovered facial region, or a combination thereof.