公开(公告)号：US20230277069A1

公开(公告)日：2023-09-07

申请号：US18011899

申请日：2022-03-03

Applicant: Google LLC

Inventor： Jiening Zhan ,  Sean Kyungmok Bae ,  Silviu Borac ,  Yunus Emre ,  Jonathan Wesor Wang ,  Jiang Wu ,  Mehr Kashyap ,  Ming Jack Po ,  Liwen Chen ,  Melissa Chung ,  John Cannon ,  Eric Steven Teasley ,  James Alexander Taylor, Jr. ,  Michael Vincent McConnell ,  Alejandra Maciel ,  Allen KC Chai ,  Shwetak Patel ,  Gregory Sean Corrado ,  Si-Hyuck Kang ,  Yun Liu ,  Michael Rubinstein ,  Michael Spencer Krainin ,  Neal Wadhwa

IPC: A61B5/0205 ,  A61B5/00

CPC classification number: A61B5/0205 ,  A61B5/0077 ,  A61B5/725 ,  A61B5/6898 ,  A61B5/7257 ,  A61B5/7278 ,  A61B5/7485 ,  A61B5/0816

Abstract: Generally, the present disclosure is directed to systems and methods for measuring heart rate and respiratory rate using a camera such as, for example, a smartphone camera or other consumer-grade camera. Specifically, the present disclosure presents and validates two algorithms that make use of smartphone cameras (or the like) for measuring heart rate (HR) and respiratory rate (RR) for consumer wellness use. As an example, HR can be measured by placing the finger of a subject over the rear-facing camera. As another example, RR can be measured via a video of the subject sitting still in front of the front-facing camera.

公开(公告)号：US20210304001A1

公开(公告)日：2021-09-30

申请号：US17215915

申请日：2021-03-29

Applicant: Google LLC

Inventor： Jack Po ,  Jiening Zhan ,  Jiang Wu ,  Silviu Borac ,  Yunus Emre

IPC: G06N3/08 ,  G06N3/04 ,  A61B5/0205 ,  G06K9/00 ,  G16H40/67 ,  G16H80/00

Abstract: A method for estimating two or more physiological signals from a subject includes steps of a) obtaining a video input in the form of a sequence of frames of image data depicting the face and optionally the chest of the subject; b) providing the video input to a multi-head neural network model trained from a set of facial video inputs from a multitude of other subjects (such video inputs optionally including the chest), wherein the model has at least two heads and is trained to predict at least two physiological signals from a video input; and c) generating with the model data representing an estimate of the two or more physiological signals of the subject. In one embodiment the physiological signals are heart rate and respiratory rate. In one embodiment the multi-head neural network model is implemented in a smartphone having a camera which is used to capture the video input.

公开(公告)号：US10089796B1

公开(公告)日：2018-10-02

申请号：US15800343

申请日：2017-11-01

Applicant: Google LLC

Inventor： Manfred Ernst ,  Silviu Borac

IPC: G06T17/00 ,  G06T19/20 ,  G06T15/04 ,  G06T7/529

Abstract: In one general aspect, a method can include combining a partition polygon and a generated texture map to form a model of a scene for rendering in three dimensions in a virtual reality space. The generating of the texture map can include projecting a Layered Depth Image sample in a partition polygon to a point in a source camera window space, projecting the point back into the partition polygon as a surface element (surfel), projecting the surfel to a surfel footprint in a target camera window space, projecting from the target camera window space to the partition polygon, sub-pixel samples included in pixels covered by the surfel footprint, projecting the sub-pixel samples from the partition polygon and into the source camera window space, and applying a color weight to each sub-pixel sample based on the location of the sample in the source camera window space.