Human Centered Computing
Download
Report
Transcript Human Centered Computing
Human-Centered Computing
Telepresence, AR, tangible interfaces
John Canny
UC Berkeley
Telepresence, AR, tangible interfaces
John Canny
• PRoPs (Personal Roving Presences) and
gesturing avatars.
• UPM - a universal planar manipulator.
• Bearable computers - turning laptops into
augmented reality systems and wearable
computers.
HCC
overview
PRoPs: Personal Roving Presences
Space Browser + Eric Paulos
Surface cruiser
HCC
overview
PRoPs: Toward being there
• A PRoP is a Personal Roving Presence.
• PRoPs seek to bridge the gap between
•
telepresence and being there.
There are two essential aspects of remote
presence that only telerobotics can provide:
Non-verbal communication, i.e. gaze, proxemics,
gesture etc. which are highly spatialized.
A gamut of human behaviors, such as chance
encounters, informal meetings, showing or being
shown, browsing, exploring etc, which are affordances
of a physical presence.
HCC
overview
Terrestrial PRoPs: Carts
• Carts are a good match to human
capabilities.
• Our strategy was to list important
physical communication skills, and
to implement as many as possible.
HCC
overview
The rest of the story:
A taxonomy of interaction behaviors
• Verbal: symbolic and prosodic
• Nonverbal communication theory (psychology,
linguistics, anthropology). Main cues are:
Gaze
Proxemics
Facial expression
Body and hand gesture
Posture
Touch
• Example: A Handshake depends on all the cues.
HCC
overview
Differences between verbal and
nonverbal communication
• Adam Kendon (83) summarizes the difference
very well:
...gesture has properties different from speech. In
particular, it employs space as well as time in the creation
of expressive forms, whereas speech can use only time. We
find, therefore, that the way information may be preserved
in speech, as compared to gesture, tends to be very
different.
HCC
overview
Why nonverbal skills are spatial
• Gaze - very sensitive to mutual
•
•
•
•
•
direction. Center-right-left.
Proxemics - Depends on distance to
others and the surroundings.
Facial expression - the least spatial,
directed by gaze.
Gesture - deictic gesture is
completely spatial, so are other
pictographic forms.
Posture - Directional.
Touch - Cant fake it without space.
HCC
overview
You need social context as well
• Communication.
• Persuasion.
• Trust-building.
• These are radically different contexts that rely
on different non-verbal cues. We need to study
telepresence in representative scenarios.
HCC
overview
The two-way street:
• PRoPs provide selective control over particular
cues. They can be turned on and off at will.
• So they provide a potentially interesting
testbed for fine-grained study of non-verbal
cues.
• We can create a rich variety of media and
study their influence on the overall perception
of the other person.
HCC
overview
The input side
• How do you achieve rich, expressive input for
•
PRoPs/avatars with today’s technology?
Cross-modal mapping from pen gesture to body
gesture. Both have symbolic and expressive
character:
HCC
overview
Future research?
• Gaze cue repair: Face to face is not perfect.
There are common breakdowns in face-to-face
communication, such as gaze aversion.
• Beyond Being There: It may be possible to
rectify gaze aversion using image processing
and/or a synthetic gaze cue.
HCC
overview
Project 2:
UPM: A universal planar manipulator
• A smart desk that can move a large number of
objects placed on it independently.
• Provides a tangible interface in both
directions for 2D layouts: floor plans,
landscapes, stage directions etc.
• Uses a small numbers of motors (4) and the
non-linearity of friction.
HCC
overview
UPM: A universal planar manipulator
• First prototype used linear voice-coils from ancient
disk drives, capable of about 2g acceleration. New
prototype should manage about 50gs.
HCC
overview
Bearable Computing (new project)
• An exploration of issues in personal, persistent
computing (augmented reality, worn interfaces)
using ordinary laptop computers.
• Avoid head-mounted displays (expensive and lowres) head-tracking, and cables.
• The approach: use optics to overlay computer
images on reality, but use laptop or pocketmounted displays.
• Testbed: Grad course in HCC next semester.
HCC
overview
An augmented reality classroom
Virtual image with notes,
questions, private chat space
Glasses to rotate
the laptop image
Physical space
HCC
overview
An AR classroom
• Students work in groups of 5-7, communicating
silently via pen or keyboard chat.
• Each group has one main note-taker, the others
add their own comments or questions to the
transcript.
• Students can mark up the group transcript, or the
lecturer’s notes. There is non-archived chat also.
• One student per group works as facilitator or TA,
posing questions to the others, and testing
understanding.
HCC
overview
Bearable computers on the go
• The laptop can serve as a “bearable” computer,
attached wirelessly to its owner.
Heads-up virtual image
Double-mirror element
in glasses
Pocket-worn portable
LCD TV
Wireless chordal keyboard or
Palmpilot in pocket for input.
Laptop in briefcase
with wireless TV
and keyboard TXs.
HCC
overview
Goals of this research
• Can we enhance attention in class with live group
chat/note-taking?
• Do we get a richer transcript with collaborative
note-taking?
• Does the head-up display make note-taking less
distracting?
• Will people do more tasks with a mobile bearable
computer because the threshold for computer use
is lower?
HCC
overview