公开(公告)号：US20230215083A1

公开(公告)日：2023-07-06

申请号：US17928874

申请日：2020-06-04

Applicant: GOOGLE LLC

Inventor： Erin Hoffman-John ,  Ryan Poplin ,  Andeep Singh Toor ,  William Lee Dotson ,  Trung Tuan Lee

IPC: G06T15/20 ,  G06T15/50 ,  G06T15/04

CPC classification number: G06T15/205 ,  G06T15/50 ,  G06T15/04

Abstract: A virtual camera captures first images of a three-dimensional (3D) digital representation of a visual asset from different perspectives and under different lighting conditions. The first images are training images that are stored in a memory. One or more processors implement a generative adversarial network (GAN) that includes a generator and a discriminator, which are implemented as different neural networks. The generator generates second images that represent variations of the visual asset concurrently with the discriminator attempting to distinguish between the first and second images. The one or more processors update a first model in the discriminator and/or a second model in the generator based on whether the discriminator successfully distinguished between the first and second images. Once trained, the generator generates images of the visual asset based on the first model, e.g., based on a label or an outline of the visual asset.

公开(公告)号：US20250047806A1

公开(公告)日：2025-02-06

申请号：US18229603

申请日：2023-08-02

Applicant: Google LLC

Inventor： Anne Menini ,  Jeya Maria Jose Valanarasu ,  Rahul Garg ,  Andeep Singh Toor ,  Xin Tong ,  Weijuan Xi

IPC: H04N7/15 ,  G06T5/00

Abstract: Methods and systems for real-time video enhancement are provided herein. A current frame of a video stream generated by a client device of a plurality of client devices participating in the video conference is identified during a video conference. An enhanced previous frame corresponding to an enhanced version of a previous frame in the video stream is identified. At least the current frame and the enhanced previous frame are provided as input to a machine-learning model. An output of the machine learning model is obtained. The output of the machine learning model indicates an enhanced current frame corresponding to an enhanced version of the current frame. The current frame is replaced with the enhanced current frame in the video stream.