Bo Wu

8131011, Mar 6, 2012

Sep 25, 2007

11/860743

Ramakant Nevatia - Pacific Palisades CA, US
Bo Wu - Mountain View CA, US

University of Southern California - Los Angeles CA

G06K 9/00

382103

A human tracking system for tracking a plurality of humans in motion, in a video of the humans in motion, includes a human detection subsystem, and a combined tracker. The human detection subsystem is configured to generate a detection output by detecting the plurality of humans in a part-based representation, in each one of a sequence of static frames in the video. The human detection subsystem is further configured to account for partial occlusion of one or more of the humans in the image. The combined tracker is configured to receive and combine the detection responses generated by the human detection subsystem, and to track the humans in response to the received detection responses and image appearance properties.

20110085702, Apr 14, 2011

Oct 8, 2010

12/901316

Ramakant Nevatia - Pacific Palisades CA, US
Chang Huang - Los Angeles CA, US
Bo Wu - Mountain View CA, US

UNIVERSITY OF SOUTHERN CALIFORNIA - Los Angeles CA

G06K 9/00

382103

Systems, methods, and computer readable storage media are described that can provide a multi-level hierarchical framework to progressively associate detection responses, in which different methods and models are adopted to improve tracking robustness. A modified transition matrix for the Hungarian algorithm can be used to solve the association problem that considers not only initialization, termination and transition of tracklets but also false alarm hypotheses. A Bayesian inference approach can be used to automatically estimate a scene structure model as the high-level knowledge for the long-range trajectory association.

20130257709, Oct 3, 2013

Jul 17, 2012

13/550791

Hayes Solos Raffle - Palo Alto CA, US
Michael Patrick Johnson - Sunnyvale CA, US
David Sparks - Cupertino CA, US
Bo Wu - Alhambra CA, US

Google Inc. - Mountain View CA

G06F 3/01

345156

This disclosure relates to proximity sensing for wink detection. An illustrative method includes receiving data from a receiver portion of a proximity sensor. The receiver portion is disposed at a side section of a head-mountable device (HMD). When a wearer wears the HMD, the receiver portion is arranged to receive light reflected from an eye area of the wearer, the proximity sensor detects a movement of the eye area, and the data represents the movement. The method includes determining that the data corresponds to a wink gesture. The method also includes selecting a computing action to perform, based on the wink gesture. The method further includes performing the computing action.

20220309278, Sep 29, 2022

Mar 29, 2021

17/216605

- Armonk NY, US
Antonio Jose Jimeno Yepes - MELBOURNE, AU
Bo Wu - Cambridge MA, US

G06K 9/00
G06K 9/62
G06K 9/46
G06N 3/04
G06N 3/08
G06F 16/735

Unsupervised learning for video classification. One or more features from one or more video clips are extracted using a spatial-temporal encoder. The one or more extracted features are processed, using a video instance discrimination task, to generate a classification label, the classification label indicating whether two of the video clips are from a same video. The one or more extracted features are processed, using a pair-wise speed discrimination task, to generate a comparison label, the comparison label indicating a relative playback speed between two given video clips. A search is performed in a video database for a video that is similar to a given video based on the comparison label.

20170061647, Mar 2, 2017

Nov 2, 2012

13/667147

Thad Eugene STARNER - Mountain View CA, US
Michael Patrick JOHNSON - Sunnyvale CA, US
Antonio Bernardo MONTEIRO COSTA - San Francisco CA, US
Bo WU - Alhambra CA, US

G09G 5/00

Methods and systems are provided for authenticating access to a wearable computing device using an authentication object, such as a hand. The wearable computing device comprises a head mountable display (HMD) and an image capture device and has access to a data profile for the authentication object. The wearable computing device provides (e.g., using the HMD) an indication for positioning the authentication object within a field of view of the image capture device. After providing the indication for positioning the authentication object, the wearable computing device receives image data from the image capture device. If the wearable computing device identifies the authentication object in the image data, for example, by matching at least a portion of the image data with data in the data profile, authentication is successful and one or more functions of the wearable computing device can be enabled.

20160057339, Feb 25, 2016

Oct 19, 2015

14/886690

- Mountain View CA, US
Sergey Brin - Palo Alto CA, US
Bo Wu - Alhambra CA, US
Michael Patrick Johnson - Sunnyvale CA, US
David Sparks - Cupertino CA, US

H04N 5/232
G06F 3/01
G02B 27/01
G06K 9/00

This disclosure relates to winking to capture image data using an image capture device that is associated with a head-mountable device (HMD). An illustrative method includes detecting a wink gesture at an HMD. The method also includes causing an image capture device to capture image data, in response to detecting the wink gesture at the HMD.

20150098620, Apr 9, 2015

Dec 16, 2014

14/571402

Bo Wu - Alhambra CA, US
Thad Eugene Starner - Mountain View CA, US
Hayes Solos Raffle - Mountain View CA, US
Yong Zhao - San Jose CA, US
Edward Allen Keyes - Mountain View CA, US

GOOGLE INC. - Mountain View CA

G06T 7/00
G06K 9/00
G01S 17/06

382103

Methods and systems are described for determining eye position and/or for determining eye movement based on glints. An exemplary computer-implemented method involves: (a) causing a camera that is attached to a head-mounted display (HMD) to record a video of the eye; (b) while the video of the eye is being recorded, causing a plurality of light sources that are attached to the HMD and generally directed towards the eye to switch on and off according to a predetermined pattern, wherein the predetermined pattern is such that at least two of the light sources are switched on at any given time while the video of the eye is being recorded; (c) analyzing the video of the eye to detect controlled glints that correspond to the plurality of light sources; and (d) determining a measure of eye position based on the controlled glints.