公开(公告)号：US20250104248A1

公开(公告)日：2025-03-27

申请号：US18977691

申请日：2024-12-11

Applicant: Snap Inc.

Inventor： Gal Dudovitch ,  Peleg Harel ,  Chia-Hao Hsieh ,  Sergei Korolev ,  Ma'ayan Mishin Shuvi

IPC: G06T7/11 ,  G06F18/21 ,  G06F18/214 ,  G06T5/70 ,  G06T11/00 ,  G06V40/10

Abstract: Methods and systems are disclosed for performing operations comprising: receiving a monocular image that includes a depiction of a whole body of a user; generating a segmentation of the whole body of the user based on the monocular image; accessing a video feed comprising a plurality of monocular images received prior to the monocular image; smoothing, using the video feed, the segmentation of the whole body generated based on the monocular image to provide a smoothed segmentation; and applying one or more visual effects to the monocular image based on the smoothed segmentation.

公开(公告)号：US12260588B2

公开(公告)日：2025-03-25

申请号：US18373554

申请日：2023-09-27

Applicant: Snap Inc.

Inventor： Sagi Katz ,  Matthias Kalkgruber

IPC: G06T7/33 ,  G06T7/593 ,  G06T7/73 ,  G06T19/00 ,  H04N13/239

Abstract: A method for correcting a bending of a flexible device is described. In one aspect, the method includes accessing feature data of a first stereo frame that is generated by stereo optical sensors of the flexible device, the feature data generated based on a visual-inertial odometry (VIO) system of the flexible device, accessing depth map data of the first stereo frame, the depth map data generated based on a depth map system of the flexible device, estimating a pitch-roll bias and a yaw bias based on the features data and the depth map data of the first stereo frame, and generating a second stereo frame after the first stereo frame, the second stereo frame based on the pitch-roll bias and the yaw bias of the first stereo frame.

公开(公告)号：US12260567B2

公开(公告)日：2025-03-25

申请号：US17973167

申请日：2022-10-25

Applicant: Snap Inc.

Inventor： Branislav Micusik ,  Georgios Evangelidis ,  Daniel Wolf

IPC: G06T7/292 ,  G06F3/01 ,  G06T7/564 ,  G06T19/00 ,  G06V20/64

Abstract: A method for carving a 3D space using hands tracking is described. In one aspect, a method includes accessing a first frame from a camera of a display device, tracking, using a hand tracking algorithm operating at the display device, hand pixels corresponding to one or more user hands depicted in the first frame, detecting, using a sensor of the display device, depths of the hand pixels, identifying a 3D region based on the depths of the hand pixels, and applying a 3D reconstruction engine to the 3D region.

公开(公告)号：US12259594B2

公开(公告)日：2025-03-25

申请号：US16711133

申请日：2019-12-11

Applicant: Snap Inc.

Inventor： John Bernard Ardisana, II ,  Yoav Ben-Haim ,  Teodor Dabov ,  Varun Sehrawat

IPC: G02C11/00 ,  G02C5/14 ,  H01R13/443 ,  H02J7/00

Abstract: Eyewear that includes a frame supporting an optical element. The frame has a first side and a second side. The eyewear also includes a temple adjacent the first side of the frame. The temple includes a first portion adjacent the frame and a second portion releasably connected to the first portion. The eyewear also includes an electrical connector embedded within the first portion of the temple. The second portion conceals the electrical connector from an exterior of the eyewear when the second portion connects to the first portion in a concealed state, and exposes the electrical connector from the exterior of the eyewear when disconnected from the first portion in an exposed state.

公开(公告)号：US20250097849A1

公开(公告)日：2025-03-20

申请号：US18969480

申请日：2024-12-05

Applicant: Snap Inc.

Inventor： Yu Jiang Tham ,  John James Robertson ,  Gerald Nilles ,  Jason Heger ,  Praveen Babu Vadivelu

IPC: H04W52/02 ,  G01S5/00 ,  H04B1/3827 ,  H04W4/029

Abstract: Systems, methods, devices, computer readable media, and other various embodiments are described for location management processes in wearable electronic devices. Performance of such devices is improved with reduced time to first fix of location operations in conjunction with low-power operations. In one embodiment, low-power circuitry manages high-speed circuitry and location circuitry to provide location assistance data from the high-speed circuitry to the low-power circuitry automatically on initiation of location fix operations as the high-speed circuitry and location circuitry are booted from low-power states. In some embodiments, the high-speed circuitry is returned to a low-power state prior to completion of a location fix and after capture of content associated with initiation of the location fix. In some embodiments, high-speed circuitry is booted after completion of a location fix to update location data associated with content.

公开(公告)号：US20250097441A1

公开(公告)日：2025-03-20

申请号：US18469256

申请日：2023-09-18

Applicant: Snap Inc.

Inventor： Yichen Wang ,  Yuechuan Li ,  Si Wang ,  Yihuan Zhou ,  Haoyun Wu ,  Junhong Nie

IPC: H04N19/436 ,  G06F21/60 ,  H04N19/136 ,  H04N19/162 ,  H04N19/40 ,  H04N19/85

Abstract: Uploading of a video file is performed by transcoding, processing and uploading portions of the video file in parallel, to reduce total processing and upload time. The processing of the video file may include applying associated augmented reality effects to a raw video recording, to generate an enhanced video recording for transmission and viewing at a recipient device. The uploaded portions of the video file may be assembled into a fragmented file format such as fMP4, in which portions of the video file are stored as fragments.

公开(公告)号：US20250095308A1

公开(公告)日：2025-03-20

申请号：US18969689

申请日：2024-12-05

Applicant: Snap Inc.

Inventor： Matan Zohar ,  Yanli Zhao ,  Brian Fulkerson ,  Riza Alp Guler

IPC: G06T19/00 ,  G06T7/20 ,  G06T7/60 ,  G06T11/00 ,  G06T17/20

Abstract: Methods and systems are disclosed for performing operations comprising: receiving a video that includes a depiction of a real-world object; generating a three-dimensional (3D) body mesh associated with the real-world object that tracks movement of the real-world object across frames of the video; determining UV positions of the real-world object depicted in the video to obtain pixel values associated with the UV positions; generating an external mesh and associated augmented reality (AR) element representing the real-world object based on the pixel values associated with the UV positions; deforming the external mesh based on changes to the 3D body mesh and a deformation parameter; and modifying the video to replace the real-world object with the AR element based on the deformed external mesh.

公开(公告)号：US20250095198A1

公开(公告)日：2025-03-20

申请号：US18965798

申请日：2024-12-02

Applicant: Snap Inc.

Inventor： Avihay Assouline ,  Itamar Berger ,  Gal Dudovitch ,  Matan Zohar

IPC: G06T7/73 ,  G06F18/214 ,  G06V20/20 ,  G06V20/40 ,  G06V20/64 ,  G06V40/10 ,  G06V40/20 ,  H04L51/04 ,  H04N21/4402

Abstract: Aspects of the present disclosure involve a system comprising a computer-readable storage medium storing a program and a method for detecting a pose of a user. The program and method include operations comprising receiving a monocular image that includes a depiction of a body of a user; detecting a plurality of skeletal joints of the body based on the monocular image; accessing a video feed comprising a plurality of monocular images received prior to the monocular image; filtering, using the video feed, the plurality of skeletal joints of the body detected based on the monocular image; and determining a pose represented by the body depicted in the monocular image based on the filtered plurality of skeletal joints of the body.

公开(公告)号：US20250093664A1

公开(公告)日：2025-03-20

申请号：US18968401

申请日：2024-12-04

Applicant: Snap Inc.

Inventor： Rory Thomas Alexander Mills ,  Ian Thomas Macken

IPC: G02B27/01 ,  G02B17/08 ,  G02B27/09

Abstract: An optical arrangement to transmit an image from an image plane to a user's eye. The arrangement providing a folded optical transmission path comprising a collimating element, having a first optical element with a first plurality of optically powered surfaces; and a second optical element comprising at least one optically powered surface. The collimating element to receive light forming the image from an image source and collimate and output the light. The optically powered surfaces having a plurality of interfaces along the folded optical path. A refractive index change at each interface is predetermined to control the direction of light passing through each interface. One surface of each of the first and the second optical elements being adjacent to one another. The adjacent surfaces having dissimilar shapes and each defining an angle with a respective other surface of the relevant optical element at opposing ends of the adjacent surfaces.

公开(公告)号：US12254412B2

公开(公告)日：2025-03-18

申请号：US18096338

申请日：2023-01-12

Applicant: Snap Inc.

Inventor： Enxu Yan ,  Sergey Tulyakov ,  Aleksei Podkin ,  Aleksei Stoliar

IPC: G06N3/082 ,  G06F17/16 ,  G06N3/04 ,  G06T7/10

Abstract: A neural network pruning system can sparsely prune neural network models using an optimizer based approach that is agnostic to the model architecture being pruned. The neural network pruning system can prune by operating on the parameter vector of the full model and the gradient vector of the loss function with respect to the model parameters. The neural network pruning system can iteratively update parameters based on the gradients, while zeroing out as many parameters as possible based a preconfigured penalty.