Pattern Recognition in Ubiquitous Computing

Nhận diện mẫu trong tính toán mọi nơi, tài liệu trình bày powerpoint do Moongu Jeon biên soạn, với nhiều nội dung mới cập nhật và hữu ích. 29 trang tài liệu với nhiều công thức và cách tính, trình bày ngắn gọn dễ hiểu

Pattern Recognition in Ubiquitous Computing

Moongu Jeon

GIST

„ Augmented reality

„ Pattern recognition problems in Ubicomp

„ Introduction to Speech Recognition

Trend of Technology Development

Ubiquitous Computing

„ Making computing an integral, invisible part

of the way people live their lives, and

available anytime anyplace

„ Computers become parts of environment , and vanish into the background

„ Integrating computers seamlessly into the world.

“Ubiquitous” in IT

„ Writing

„ The first IT freed information from the limits of individual memory

„ Books,magazines, newspapers, street signs,

billboards, shop signs, candy wrappers which are parts of the environment- ubiquitous

„ Current silicon-based IT

„ Huge number of computers and communication devices is far from having becoming of the

environment

Two Issues in Ubicomp

„ Location

„ If a computer merely knows what room it is in,

it can adapt its behavior in significant ways

without requiring even a hint of artificial

intelligence

„ Scale (size)

„ Tabs (inch-scale machine): Post-It notes

„ Pads (foot-scale): book or magazine

„ Boards (yard-scale): black (or bulletin) board

„ Identify and keep track

of users or objects, and

do more tasks

„ Roy Want, PARC – tab incorporating a small display

„ Serves simultaneously

as an active badge, calendar, diary

„ Scrap computer (analogous to scrap paper)

„ Can be grabbed and used anywhere.

„ Have no individual identity.

Other Issues Ubicomp

„ Cheap, low-power hardware

components.

„ A network that ties them all together.

„ Software for screens and pens

„ Applications.

„ Privacy

„ Computational methods

Augmented Reality

„ The opposite approach from virtual reality.

„ VR encloses people in an artificial world using computers

„ AR augments objects in the real world using computers

„ Examples

„ Digitaldesk (Wellner 1993)

„ KARMA (Feiner 1993)

„ Flatland (Mynatt 1999) –augmented whiteboard

„ UbiTV, MRWindow, ARTable (Woo 2006)

Mobile AR

„ What is AR?

„ To enhance the user’s

perception of and interaction with the real world through supplementing the real world with 3D virtual objects that appear to coexist in the

same space as the real

Mobile AR (개념도)

ubiTV

사용자 A

특정 사용자와의 선택적 콘텐츠 공유

■ ▶ ↑ ↓

ubiTV

프로파일에 따라 개인화된 콘텐츠 증강 선택적 콘텐츠 공유

Mobile AR in U-Space

Future of Ubicomp

„ A key concept of ubicomp is to use technology to create a calmer

environment (Weiser 1998).

„ Computer technology should serve

humans as environment that does not occupy much of human attention,

and should serve humans calmly not consuming human effort

„ Collection of user’s data using wireless sensors

„ Recognition of user’s behavior pattern

„ Object or image recognition to get the

augmented data in U-space.

„ Speech recognition

Classifier design

System evaluation pattern

Speech Recognition

„ Isolated word recognition (IWR)

„ Continuous speech recognition (CSR)

„ Speaker-dependent recognition

„ Speaker-independent recognition

„ Dynamic time warping (DTW) - DP

„ finds an optimal match between two sequences

of feature vectors which allows for streched

and compressed sections of the sequence

Symmetrical DTW

Dynamic Time Warping

„ Global constraints

„ Local constraints

„ Monotonicity - matching paths cannot go

backwards in time

„ End point constraints

„ Starts at (0,0) and ends at (Ι,J) and whose first transition is to the node (1,1)

„ The cost for the transitions

„ Euclidean distance

DTW Global Constraint

DTW Local Constraints

„ Sakoe and Chiba local constraints

Example

Test Words

Feature Selection

„ r(j), j=1,…,J -Æ reference pattern

„ t(i), i=1,…,Ι -Æ test pattern

„ xj(n), n=0,…511 -Æ samples for the

jth frame of the reference pattern.

„ Taking DFT

„ Xi(m)=∑xi(n)exp(-j2 πmn/512)/5121/2, m=0,…,511

„ Use the first 50 DFT coeff as features

„ r(j)=[Xj(0) Xj(1) … Xj(49)]T, j=1,…J