Browse Prior Art Database

GESTURE-BASED CURSOR CONTROL IN VIDEO CONFERENCE MEETING

IP.com Disclosure Number: IPCOM000250605D
Publication Date: 2017-Aug-08
Document File: 8 page(s) / 829K

Publishing Venue

The IP.com Prior Art Database

Related People

Zhaocai Wang: AUTHOR [+3]

Abstract

Techniques are presented herein by which a video camera recognizes a user's body region and hand gesture so that the user is able to control a cursor on video conference screen simply by waving his hand, without tedious verbal description or using a physical computer mouse or other similar pointing device. Using this method, other people in the same video conference room or remote users in the same video conference meeting can clearly see where on the screen the active speaker is referring.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 40% of the total text.

Copyright 2017 Cisco Systems, Inc. 1

GESTURE-BASED CURSOR CONTROL IN VIDEO CONFERENCE MEETING

AUTHORS: Zhaocai Wang

Yasi Xi Kent Chen

CISCO SYSTEMS, INC.

ABSTRACT

Techniques are presented herein by which a video camera recognizes a user's body

region and hand gesture so that the user is able to control a cursor on video conference

screen simply by waving his hand, without tedious verbal description or using a physical

computer mouse or other similar pointing device. Using this method, other people in the

same video conference room or remote users in the same video conference meeting can

clearly see where on the screen the active speaker is referring.

DETAILED DESCRIPTION

Presented herein are techniques to improve the user experience during video

conference meetings. Consider the following two scenarios:

1) A group of people are having a meeting in a video conference room and one of

them is sharing his laptop desktop onto the video conference display screen. If someone

else in the room, e.g. Alex, wants to speak about some keyword or some area in the video

conference display screen, he may say

• "Please pay attention to the project estimation in the first line of the second

paragraph ..."

or

• "Please look at the burndown chart of this project on the left top of the screen ..."

2) Bob is having an online meeting in a video conference room with remote

attendees. One of the remote attendees is sharing his desktop on the video conference

screen. Bob wants to speak about some keyword or some area in the video conference

screen, he may say

Copyright 2017 Cisco Systems, Inc. 2

• "Please pay attention to the project estimation in the first line of the second

paragraph ..."

or

• "Please look at the burndown chart of this project on the left top of the screen ..."

In both scenarios, it is cumbersome because Alex/Bob has spoken repeatedly but

the others may not immediately know what he is referring to on the video conference screen.

Sometimes, words are weak in terms of visual perception.

Gestures can be more natural and visual than words. Presented herein is a method,

of gesture-based cursor control for a video conference meeting.

Certain video conference systems, such as Cisco System’s Telepresence system,

have speaker tracking capabilities, such as Cisco System’s SpeakerTrack™ 60 system.

When someone speaks, the video camera focuses on and zooms in to that speaker. Thus,

the speaker is accurately captured by the camera.

Meanwhile, the following technologies are becoming more and more mature.

• Human body region extraction. For example, human body region extraction from

photos, extracting moving people from Internet videos.

• Hand gesture recognition. For example, hand gesture recognition using a real-time

tracking method and hidden Markov models, orientation histograms for hand

gesture recognition, gesture and fingers recognition.

By leveraging the existing technologies above, a video conference endpoint camera

recognizes an active speaker's body reg...