-
公开(公告)号：US12182722B2
公开(公告)日：2024-12-31
申请号：US18213145
申请日：2023-06-22
Applicant: Snap Inc.
Inventor： Sergey Tulyakov , Sergei Korolev , Aleksei Stoliar , Maksim Gusarov , Sergei Kotcur , Christopher Yale Crutchfield , Andrew Wan
IPC: G06N3/088 , G06F18/21 , G06F18/214 , G06N3/045 , G06N3/08 , G06V10/764 , G06V10/774 , G06V10/778 , G06V10/82
Abstract: A compact generative neural network can be distilled from a teacher generative neural network using a training network. The compact network can be trained on the input data and output data of the teacher network. The training network train the student network using a discrimination layer and one or more types of losses, such as perception loss and adversarial loss.
-
公开(公告)号：US20240427427A1
公开(公告)日：2024-12-26
申请号：US18829090
申请日：2024-09-09
Applicant: Snap Inc.
Abstract: A system for hand tracking for an Augmented Reality (AR) system. The AR system uses a camera of the AR system to capture tracking video frame data of a hand of a user of the AR system. The AR system generates a skeletal model based on the tracking video frame data and determines a location of the hand of the user based on the skeletal model. The AR system causes a steerable camera of the AR system to focus on the hand of the user.
-
公开(公告)号：US12177559B2
公开(公告)日：2024-12-24
申请号：US18242016
申请日：2023-09-05
Applicant: Snap Inc.
Inventor： Victor Shaburov , Alexander Mashrabov , Grigoriy Tkachenko , Ivan Semenov
IPC: H04N23/63 , G06V40/16 , H04N5/265 , H04N5/272 , H04N23/611
Abstract: Systems and methods for providing personalized videos are provided. An example method includes receiving preprocessed videos including a target face and facial expression parameters of the target face, modifying the preprocessed videos to generate one or more personalized videos by replacing the target face with a source face, where the source face is modified to adopt the facial expression parameters of the target face, providing a user interface enabling a user to share at least one personalized video of the one or more personalized videos with a further user of a further computing device, determining that an application to be used to share the personalized video does not allow auto-play of the personalized video in a video format, in response the determination, exporting the personalized video of the one or more personalized videos into an image file, and sharing the image file via the application.
-
公开(公告)号：US12176005B2
公开(公告)日：2024-12-24
申请号：US18243487
申请日：2023-09-07
Applicant: Snap Inc.
Inventor： Nathan Kenneth Boyd , Devin Doty
IPC: H04N5/93 , G06V20/40 , G11B27/00 , G11B27/031 , H04N21/431 , H04N21/472
Abstract: Systems and methods provide for advanced video editing techniques using sampling patterns. In one example, a computing device can receive a selection of a clip of a video and a sampling pattern. The computing device can determine a respective number of frames to sample from the clip for each interval of time over a length of time for a new clip. For example, the computing device can determine a function corresponding the pattern that relates time and the number of frames to sample, a histogram corresponding to the pattern, or a definite integral corresponding to the pattern, among other approaches. The computing device can extract these numbers of frames from the clip and generate the new clip from the extracted frames. The computing device can present the new clip as a preview and send the new clip to other computing devices.
-
公开(公告)号：US12175623B2
公开(公告)日：2024-12-24
申请号：US18373563
申请日：2023-09-27
Applicant: Snap Inc.
Inventor： Bernhard Jung , Edward Lee Kim-Koon
Abstract: A method for adjusting an over-rendered area of a display in an AR device is described. The method includes identifying an angular velocity of a display device, a most recent pose of the display device, previous warp poses, and previous over-rendered areas, and adjusting a size of a dynamic over-rendered area based on a combination of the angular velocity, the most recent pose, the previous warp poses, and the previous over-rendered areas.
-
Patent Agency Ranking
公开(公告)号：US12175613B2
公开(公告)日：2024-12-24
申请号：US18375693
申请日：2023-10-02
Applicant: Snap Inc.
Inventor： Kyle Goodrich , Samuel Edward Hare , Maxim Maximov Lazarov , Tony Mathew , Andrew James McPhee , Daniel Moreno , Wentao Shang
IPC: G06T19/00 , G06F3/01 , G06F3/04847 , G06F3/04883 , G06T19/20 , H04L51/10
Abstract: Aspects of the present disclosure involve a system comprising a computer-readable storage medium storing at least one program and method for performing operations comprising: receiving, by one or more processors that implement a messaging application, a video feed from a camera of a user device; detecting, by the messaging application, a face in the video feed; in response to detecting the face in the video feed, retrieving a three-dimensional (3D) caption; modifying the video feed to include the 3D caption at a position in 3D space of the video feed proximate to the face; and displaying a modified video feed that includes the face and the 3D caption.
-
公开(公告)号：US12175605B2
公开(公告)日：2024-12-24
申请号：US17700733
申请日：2022-03-22
Applicant: Snap Inc.
Inventor： Edmund Graves Brown , Benjamin Lucas , Jonathan M. Rodriguez, II , Richard Zhuang
Abstract: Content is displayed to a user of augmented reality device. In response to receiving an indication of an increased level of risk, the degree of content being displayed to the user is reduced. The indication of increased level of risk may be generated by or received from an associated transportation device. The adjustment of the display of the degree of content may include moving one or more content elements out of a central field of view of the augmented reality device, reducing the size or visual characteristics of a content element, or eliminating a content element from the display.
-
公开(公告)号：US12175570B2
公开(公告)日：2024-12-24
申请号：US17657286
申请日：2022-03-30
Applicant: Snap Inc.
Inventor： Michael Kozakov , Avie Seegobin , Mark Anthony Cabuena
Abstract: Systems, methods, and computer readable media for customizable avatar generation system, where the methods include accessing text data, processing, using at least one processor, the text data to determine first characteristics of the text data, selecting a personalized avatar of a plurality of personalized avatars for the text data based on matching the first characteristics with second characteristics of the plurality of personalized avatars, generating a customized avatar based on the text data and the selected personalized avatar, and causing the customized avatar to be displayed on a display of a computing device.
-
公开(公告)号：US20240420407A1
公开(公告)日：2024-12-19
申请号：US18211149
申请日：2023-06-16
Applicant: Snap Inc.
Inventor： Evangelos Ntavelis , Kyle Olszewski , Aliaksandr Siarohin , Sergey Tulyakov
Abstract: Systems and methods for generating static and articulated 3D assets are provided that include a 3D autodecoder at their core. The 3D autodecoder framework embeds properties learned from the target dataset in the latent space, which can then be decoded into a volumetric representation for rendering view-consistent appearance and geometry. The appropriate intermediate volumetric latent space is then identified and robust normalization and de-normalization operations are implemented to learn a 3D diffusion from 2D images or monocular videos of rigid or articulated objects. The methods are flexible enough to use either existing camera supervision or no camera information at all—instead efficiently learning the camera information during training. The generated results are shown to outperform state-of-the-art alternatives on various benchmark datasets and metrics, including multi-view image datasets of synthetic objects, real in-the-wild videos of moving people, and a large-scale, real video dataset of static objects.
-
公开(公告)号：US20240420382A1
公开(公告)日：2024-12-19
申请号：US18819587
申请日：2024-08-29
Applicant: Snap Inc.
Inventor： Tejas Bahulkar
IPC: G06T11/00 , G06F3/04842 , G06F3/0488 , G06F3/16 , G06V20/20 , G06V20/68
Abstract: Example systems, devices, media, and methods are described for curating and presenting a contextual overlay that includes graphical elements and virtual elements in an augmented reality experience. A contextual overlay application implements and controls the capturing of frames of video data within a field of view of the camera. The image processing system detects, in the captured frames of video data, one or more food items in the physical environment. Detecting food items may involve computer vision and machine-trained classification models. The method includes retrieving data associated with the detected food item, curating a contextual overlay based on the retrieved data and a configurable profile, and presenting the contextual overlay on the display.
-
-
-
-
-
-
-
-
-
Search Results
5367
records
(took 0.04 seconds)
