블로그 이미지
Leeway is... the freedom that someone has to take the action they want to or to change their plans.
maetel

Notice

Recent Post

Recent Comment

Recent Trackback

Archive

calendar

1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30
  • total
  • today
  • yesterday

Category

2011. 10. 19. 02:26 Method/Sound
Miller Puckette [The Theory and Technique of Electronic Music]
http://crca.ucsd.edu/~msp/techniques.htm


Foreword by Max Mathews


http://en.wikipedia.org/wiki/Max_Mathews

"block diagram compilers with graphical interfaces"

> Max by Miller Puckette
- "1st graphical compiler program"
- http://en.wikipedia.org/wiki/Max_(software)
- "data-flow system"

cf. 검색병이 도져 우연히 찾은 자료: Jeffrey Hass (Indiana University)'s Introduction to Computer Music: Volume One
(사운드, 음파에 대해 처음으로 이해해 보려는 나에게 안성맞춤. 설명용 그래픽 애니메이션이 정성스럽고, 무엇보다 최대 미덕은 바로 짧은 분량. 핵심 요약 노트 같다.)


Preface


Miller Puckette http://www-crca.ucsd.edu/~msp/

ref.
John Strawn, F. Richard Moore, Digital audio signal processing: an anthology (Volume 1 of Computer music and digital audio series)


 

'Method > Sound' 카테고리의 다른 글

MaxScore  (0) 2008.07.01
소리의 3요소  (0) 2008.05.28
3-D Sound for Virtual Reality and Multimedia  (0) 2008.05.27
4. Synthesis Fundamentals  (0) 2008.05.26
Understanding Basic Music Theory  (0) 2008.05.26
posted by maetel
2011. 8. 25. 11:34 Computer Vision
The Discipline of Machine Learning


Tom M. Mitchell (July 2006), School of Computer Science Carnegie Mellon University Pittsburgh


Tom M. Mitchell http://www.cs.cmu.edu/~tom/



'Computer Vision' 카테고리의 다른 글

tensor  (1) 2011.07.07
Gonzalez & Woods [Digital Image Processing]  (0) 2011.06.16
Seam Carving  (0) 2011.01.17
Jordan & Bishop "Neural Networks"  (0) 2010.12.14
Michael I. Jordan, "Generic constraints on underspecified target trajectories"  (0) 2010.12.11
posted by maetel
2011. 8. 22. 09:53 Footmarks
올해로 세번째 개최되는 The 3rd International Symposium on Brain and Cognitive Engineering (뇌공학 국제심포지엄)이 아래와 같이 개최되오니 재학생 및 관심있는 여러분의 많은 참여 바랍니다.

일시: 2011년 8월 30일(화) ~ 31일(수)
장소: 고려대학교 자연계캠퍼스 하나스퀘어
주최: 고려대학교 WCU 뇌공학연구사업단
후원: 고려대학교, 교육과학기술부, 한국연구재단
문의: http://brain.korea.ac.kr/bcesymposium2011


posted by maetel
2011. 7. 7. 23:20 Computer Vision
http://en.wikipedia.org/wiki/Tensor

http://mathworld.wolfram.com/Tensor.html
 
kipid@다음: 텐서란? 

Joseph C. Kolecki @NASA: An Introduction to Tensors for Students of Physics and Engineering





posted by maetel
2011. 6. 16. 02:05 Computer Vision
Digital Image Processing, 3rd ed.
Rafael C. Gonzalez & Richard E. Woods
Prentice Hall, 2008

google books: http://books.google.com/books?id=8uGOnjRGEzoC
official website: http://www.imageprocessingplace.com/

Rafael C. Gonzalez 
http://www.eecs.utk.edu/people/faculty/emeritus/gonzalez/main


1. Introduction
 
1.1 What is digital image processing?

digital image processing; processing digital images by means of a digital computer

picture elements = image elements = pels = pixels

image processing -> image analysis -> computer vision



- low-level process: both inputs and outputs are images
  eg. image preprocessing to reduce noise, contrast enhancement, image sharpening
- mid-level process:  inputs generally are images, but its outputs are attributes like edges, contours, identity of objects
  eg. segmentation, description of the objects, classification (recognition)
- higher-level process: recognized objects performs the cognitive functions associated with vision


   
1.2 The origins of digital image processing

halftone 망판(網版)화
http://terms.naver.com/entry.nhn?docId=270950
http://terms.naver.com/entry.nhn?docId=64830

 
http://en.wikipedia.org/wiki/Bartlane_cable_picture_transmission_system

http://en.wikipedia.org/wiki/Punched_tape


"modern digital computer"
with the introduction by John von Neumann of two key concepts: (1) a memory (2) conditional branching, which are the foundation of a CPU
+
mass storage & display systems
=> digital image processing


 
> the birth of digital image processing

- space probe

Work on using computer techniques for improving images from a space probe began at the Jet Propulsion Laboratory (Pasadena, California) in 1964 when pictures of the moon transmitted by Ranger 7 were processed by a computer to correct various types of image distortion inherent in the on-board television camera.

- medical diagnosis
Tomography consists of algorithms that use the sensed data to construct an image that represents a "slice" through the object. Motion of the object in a direction perpendicular to the ring of detectors produces a set of such slices, which constitute a three-dimensional rendition of the inside of the object. Tomography was invented independently by Sir Godfrey N. Hounsfield and Professor Allan M. Cormack, who shared the 1979 Nobel Prize in Medicine for their invention.



computerized axial tomography (CAT) http://en.wikipedia.org/wiki/Computerized_axial_tomography
서울대학교병원: 컴퓨터 단층 촬영 (Computed Tomography)


> applications of digital image processing 
   (1) for human interpretation
   (2) for machine perception


1.3 Examples of fields that use digital image processing

http://en.wikipedia.org/wiki/Electromagnetic_spectrum


http://en.wikipedia.org/wiki/Gamma_ray

positron emission tomography (PET) http://en.wikipedia.org/wiki/Positron_emission_tomography
양전자방출 단층촬영술 http://100.naver.com/100.nhn?docid=281698

http://en.wikipedia.org/wiki/X_ray

http://en.wikipedia.org/wiki/Ultraviolet

fluorescence microscopy http://en.wikipedia.org/wiki/Fluorescence_microscope
형광현미경 http://terms.naver.com/entry.nhn?docId=434197

http://en.wikipedia.org/wiki/Visible_light

light microscopy http://en.wikipedia.org/wiki/Light_microscopy#Optical_microscopy

http://en.wikipedia.org/wiki/Infrared

http://en.wikipedia.org/wiki/Thematic_map

http://landsat.gsfc.nasa.gov/images/

http://en.wikipedia.org/wiki/Multispectral_imaging

http://en.wikipedia.org/wiki/Defense_Meteorological_Satellite_Program

National Oceanic and Atmospheric Administration (NOAA) http://www.noaa.gov

http://en.wikipedia.org/wiki/Structured_light

Perceptics Corporation http://www.perceptics.com/

radar = radio detection and ranging  http://en.wikipedia.org/wiki/Radar

사이언스올 과학용어사전: 전파(radio wave) 

http://en.wikipedia.org/wiki/Magnetic_resonance_imaging

http://en.wikipedia.org/wiki/Microwaves


acoustic imaging
electron microscopy
synthetic (computer-generated) imaging


psi 평방 인치당 파운드(pounds per square inch: 타이어 등의 압력을 나타낼 때 씀) http://en.wikipedia.org/wiki/Pounds_per_square_inch

http://en.wikipedia.org/wiki/Ultrasound_imaging

http://en.wikipedia.org/wiki/Electron_microscope

transmission electron microscope (TEM) http://en.wikipedia.org/wiki/Transmission_electron_microscopy
투과전자현미경 (透過電子顯微鏡) http://terms.naver.com/entry.nhn?docId=432217

scanning electron microscope (SEM) http://en.wikipedia.org/wiki/Scanning_electron_microscope
주사전자현미경 (走査電子顯微鏡) http://terms.naver.com/entry.nhn?docId=422329


http://en.wikipedia.org/wiki/Fractal

tiling http://en.wikipedia.org/wiki/Tessellation

http://en.wikipedia.org/wiki/3d_modeling

http://en.wikipedia.org/wiki/3D_visualization


1.4 Fundamental steps in digital image processing

image acquisition,
image enhancement,
image restoration,
color image processing,
wavelets (-> image data compression / pyramidal representation),
compression,
morphological processing,
segmentation,
(boundary/regional) representation & description (feature selection),
recognition 


1.5 Components of an image processing system

- sensing - physical device (image sensor) + digitizer

- specialized image processing hardware - digitizer + ALU
http://en.wikipedia.org/wiki/Arithmetic_logic_unit

- general-purpose computer

- software - a package of specialized modules 

- mass storage - computer memory / frame buffers (=> zoom, scroll, pan)
http://en.wikipedia.org/wiki/Frame_buffer













 
posted by maetel
2011. 6. 13. 14:12 Footmarks
주제: Optoelectronics for Brain and Cognitive Engineering
연사: 한재호 교수 (고려대학교 뇌공학과)
일시: 2011년 06월 13일(월) 오후 5시
장소: 자연계캠퍼스 아산이학관 633호
주최: 고려대학교 정보통신대학 뇌공학과
후원: 고려대학교 정보통신대학, 뇌인지과학 연계전공, WCU 뇌공학연구사업단, 뇌공학연구소
문의: 02)3290-5920


초록: Optoelectronics for Brain and Cognitive Engineering

In the first half of this talk, an introduction to various cutting-edge optoelectronic imaging and sensing technologies for high-resolution brain imaging will be overviewed. In the latter half, recent studies that have been performed will be provided mainly focusing on the optical coherence tomography (OCT). OCT has emerged as a promising imaging modality that can provide non-invasive high-resolution tomographic imaging in real-time. Novel fiber optic probes, image processing methods, near infrared sources for sensing, and surgical robot applications that could make the OCT a practical system for a high-resolution endoscopic imaging will be presented.



http://en.wikipedia.org/wiki/Optoelectronics

 

 
posted by maetel
2011. 6. 9. 20:36 Literacy
The Psychology of Art and the Evolution of the Conscious Brain
Robert L. Solso
The MIT Press, 2003
google books: http://books.google.com/books?id=z_cQJgAACAAJ


APS(Association for Psychological Science): Robert L. Solso (1933-2005)  


- To develop connections between cognitive psychology and the related fields of anthropology, computer science, education, linguistics, neuroscience, and philosophy

Stephen E. Palmer http://psychology.berkeley.edu/faculty/profiles/spalmer.html
Visual Perception and Aesthetics Lab @UCB http://socrates.berkeley.edu/~plab/

http://en.wikipedia.org/wiki/Vision_science

[Cognition and the Visual Arts] The MIT Press, 1994 - google books: http://books.google.com/books?id=zaQCKizo0_8C



Preface

“Art and cognition, and the brain, and consciousness, and evolution have all stood as complex mirrors, all reflecting and amplifying each other.”

In searching for a rational connection between consciousness and art, it was necessary to examine the evolution of the human brain and cognition. Out of these scientific explorations, I have developed a new model describing the evolution of consciousness and its relationship to the emergence of art.


conscious AWAREness

We have a pretty good idea, for example, as to when and how the human brain evolved and when early art emerged, and we have a sound understanding of the workings of the sensory-cognitive system. With this knowledge in hand, it is propitious to consider the evolution of the human brain and the emergence of AWAREness, as they might be related to art. As the brain increased in size and capacity during the upper Pleistocene, additional components of consciousness were added or developed. People became more AWARE in the sense that they were more cognizant, not only of a world that existed in contemporaneous actuality, but of a world that could be imaged. That change took humankind on a wondrous voyage. Men and women could imagine nonpresent things such as what might be behind a bush, where fresh water might be found, and what a nonpresent bull might look like. While other animals had some forms of consciousness, the visionary aptitude of humans to extend consciousness beyond responding to moment-to- moment sensory experiences was spinning into new possibilities previously unseen on this earth. Equipped with expanded conscious AWAREness, people first created art and then technology. The beginning of art is a clear manifestation of the brain’s capacity for imaginative behavior.

All factors—brain, anthropology, cog- nition, and art—were tied together by human consciousness.



Introduction: Art . . . a Tutorial












 
posted by maetel
2011. 6. 8. 12:43 Footmarks
초청: Neural Interface Lab(Director: 김성필)
일시: 2011년 6월 8일 (수), 오후 5시
장소: 고려대학교 자연계캠퍼스 미래융합기술관 111호
연사: 송주현 교수
        (Department of Cognitive, Linguistics and Psychological Sciences, Brown University)
제목: How do perception, cognition, and action interact in a complex visual environment?

 



posted by maetel
2011. 5. 28. 12:33 Footmarks
2011-05-26 나무의 늦은 일곱 시@숨도

연사: 김항 
전공: 신문방송학 학사 -> 표상문화론 박사


- 88 올림픽 이후

> 해외여행 자유화. 검열 완화
> 포스트 담론 (정치경제학 -> 문화론)
> "빠른 시간 안에 많은 사람들이 동시적으로 향유한다는 것으로 (일상의) 문화가 바뀜"
ref.

- 김우창 <포스트 모더니즘과 ~ > - '김포공항', '압구정'
- 박명신
> "집합적 주체가 아닌 개인으로서 세계와 마주하기"
cp.
- 마르크스: 삶이 관념을 결정한다 (유물론적 역사관)
- 자연주의




김항: 문화이론이란?

"사람의 삶으로/손으로 구성되지 않은 것은 없다."
"삶의 신체적 감각"
"나(고유한 존재)의 신체에 각인된 다양한 유산을 계보적으로 추적하기 
"주체/주어 없는 복수의 시간성을 복원/구축하기"

ref. 김수영 <거대한 뿌리>(1964)




이유?
- "진리/규범의 탐구가 아니라, 진리/규범을 만들어 내는 사회적 과정의 불투명함과 수치스러움"

ref. 발터 벤야민 <역사 철학 폐쇄> 중 '역사의 개념에 관하여' 
- "미래에 등을 돌린 채 과거로부터 불어오는 바람에 두 눈을 뜨고 온몸을 내맡김"
- 폐허 속에 쌓여 가는 잔해들 하나하나에 주목
 
 
김항: '환승 = 자신이 지금까지 믿고 살아 왔던 것들을 멈춤'




ref. 

- 존 스토리 <문화연구와 문화이론> 현실문화연구
- 프란츠 파농 <대지의 저주받은 자들> 그린비 / <검은 피부, 하얀 가면> 인간사랑
- 다케우치 요시미 <일본과 아시아> 소명 / <루신 연구> 그린비
- 정수진 <무형문화재의 탄생> 역사비평사
- 김홍중 <마음의 사회학> 문학동네

eg. 식민지: 인간에 대한 종류의식의 파괴










posted by maetel
2011. 5. 22. 23:01 Context
Gamer Psychology - Designing Hit Video Games (May 2011)
http://www.situatedresearch.com/webinars/webinar13.html

waterfall design
http://en.wikipedia.org/wiki/Waterfall_model

quality assurance (QA) vs. usability

> usability research
- natural environment (with invisible cameras)
- activity theory http://en.wikipedia.org/wiki/Activity_theory
- Ethno-methodology http://en.wikipedia.org/wiki/Ethnomethodology
- Grounded-theory (finding patterns) http://en.wikipedia.org/wiki/Grounded_theory
- emergent behavioral patterns http://en.wikipedia.org/wiki/Emergent_behavior
- video analysis


> gaming companies' concerns
- codes
- game play
- interface
- story lines



Game play is a psychological experience.
: ego, paranoia, delusion, self-destructive behaviors, ...

What keeps players playing your games?




Sid Meier http://en.wikipedia.org/wiki/Sid_Meier

GDC http://en.wikipedia.org/wiki/Game_Developers_Conference

artificial intelligence (AI)



http://en.wikipedia.org/wiki/MMOG
MMOG's are the future of gaming.

> Engagement

> Interaction Design
Think in terms of the tasks, not in terms of the system.


http://www.situatedresearch.com/CogTech14-2-15-1.pdf





posted by maetel
2011. 5. 22. 22:52 Context
클라우드 서비스의 3가지 본질적 속성 [정보통신정책연구원 2011.05.16]

최근 들어 IT 산업에서 클라우드 컴퓨팅이 차지하는 비중이 높아지고 있다. 그럼에도 불구하고, 클라우드 컴퓨팅에 대한 개념과 내용이 불확실한 것이 현실이다. 본고에서는 클라우드 컴퓨팅과 관련한 논의가 가진 개념과 범주의 불확실성을 극복하기 위해 클라우드 컴퓨팅의 가장 본질적인 것으로 이해될 수 있는 속성을 도출하였다. 하드웨어 통합, 데이터 이전, 아키텍처의 세 가지 이슈를 중심으로 클라우드 컴퓨팅의 서비스와 전개 모델과 상관없이 클라우드 컴퓨팅 논의에 포함되어야 하는 최소한의 공통점인 본질적 속성에 대해 정리하였다.

2006년 구글 회의에서 처음으로 등장한 ‘클라우드 컴퓨팅(cloud computing)’이라는 용어는 최근 2~3년 사이에 IT산업의 중요한 화두로 자리매김하였다. ‘클라우드 컴퓨팅(cloud computing)’은 가트너가 선정한 10대 전략에서 2010년에 이어 2011년에도 1위를 차지하였으며, 구글, IBM, MS 등 글로벌 IT 기업들이 클라우드 컴퓨팅을 핵심 사업으로 삼고 있다. 클라우드 컴퓨팅에 대한 이 같은 뜨거운 관심을 반영하여 우리 정부에서도 이를 육성하기 위한 움직임을 보이고 있다. 이러한 움직임의 하나로 2009년 말에 국내 클라우드 컴퓨팅 산업을 육성하기 시작하여 2014년 세계시장 점유율 10%에 이르는 클라우드 컴퓨팅 강국 실현을 목적으로 하는 ‘범정부 클라우드 활성화 종합계획’을 행정안전부, 지식경제부, 방송통신위 등의 3개 부처가 공동으로 수립하였다.



'Context' 카테고리의 다른 글

Macluhan in Europe 2011  (0) 2012.02.23
Situated Research - Gamer Psychology  (0) 2011.05.22
Sam Gosling's Snoop  (0) 2011.05.22
Mixed Reality general  (0) 2009.09.21
[유비유넷] 공동세미나 - Future UI - 증강현실  (0) 2008.11.25
posted by maetel
2011. 5. 22. 22:49 Context
http://snoopology.com/

http://www.newsweek.com/2008/06/22/you-are-what-you-keep.html

 
posted by maetel
2011. 5. 17. 23:42 Literacy
The New Media Reader
Edited by Noah Wardrip-Fruin and Nick Montfort 
The MIT Press, February 2003

google books: http://books.google.com/books?id=DQYXoRx9CcEC


Noah Wardrip-Fruin  http://games.soe.ucsc.edu/people/noah-wardrip-fruin

Nick Montfort  http://nickm.com




A User's Manual


http://jordanmechner.com/karateka/
http://en.wikipedia.org/wiki/Karateka_(video_game)
http://www.youtube.com/watch?v=GHNT7mR-8d0

cf. http://en.wikipedia.org/wiki/Carmen_Sandiego





 
posted by maetel
2011. 5. 3. 11:31 Literacy
http://en.wikipedia.org/wiki/Italo_Calvino

http://des.emory.edu/mfp/calvino/

http://www.italo-calvino.com/


- video
The distance from the moon - Italo Calvino http://youtu.be/EZ9cEZhiGPw
JOHN TURTURRO reads a short story by Italo Calvino http://youtu.be/bqHU9oh0hMA


posted by maetel
2011. 5. 3. 09:46 Method
posted by maetel
2011. 5. 1. 20:39 Footmarks
posted by maetel
2011. 2. 10. 13:49 Computation
웹 업로드용으로 png가 부담스러워서 jpg로 바꾸고자 한다.

ref. 

터미널에서 다음과 같이 치면 된다고.

$$ defaults write com.apple.screencapture type jpg

물론 jpg 자리에 tiff나gif나 taga 등을 넣을 수 있다. 

그리고 재부팅해야 적용된단다.


cf.

'Computation' 카테고리의 다른 글

안드로이드 내부 구조 Day #3  (0) 2011.10.26
안드로이드 내부구조 #1  (0) 2011.10.24
Matthias Felleisen et al. <How to Design Programs>  (0) 2010.09.26
Octave on Mac OS X  (1) 2009.08.12
Cloud Computing 클라우드 컴퓨팅  (0) 2009.07.03
posted by maetel
2011. 1. 17. 01:21 Computer Vision
Shai Avidan, Ariel Shamir, Seam Carving for Content-Aware Image Resizing, SIGGRAPH, 2007





posted by maetel
2010. 12. 14. 20:02 Computer Vision
Michael I. Jordan & Christopher M. Bishop, "Neural Networks", In Tucker, A. B. (Ed.) CRC Handbook of Computer Science, Boca Raton, FL: CRC Press, 1997.
download: http://www.cs.berkeley.edu/~jordan/papers/crc.ps







1. Introduction


Neural network methods have had their greatest impact in problems where statistical issues dominate and where data are easily obtained.

"conjunction of graphical algorithms and probability theory":
A neural network is first and foremost a graph with patterns represented in terms of numerical values attached to the nodes of the graph and transformations between patterns achieved via simple message-passing algorithms. Many neural network architectures, however, are also statistical processors, characterized by making particular probabilistic assumptions about data.


Based on a source of training data, the aim is to produce a statistical model of the process from which the data are generated so as to allow the best predictions to be made for new data.

statistical modeling - density estimation (unsupervised learning), classification & regression

density estimation ("unsupervised learning")
: to model the unconditional distribution of data described by some vector
- to train samples and a network model to build a representation of the probability density 
- to label regions for a new input vector
 
classification & regression ("supervised learning")
: to distinguish between input variables and target variables
- to assign each input vector to one of classes and target variables to class labels 
-> estimation of conditional densities from the joint input-target space



2. Representation

2.1 Density estimation

To form an explicit model of the input density

Gaussian mixture distribution







2.2 Linear regression and linear discriminants






posted by maetel
2010. 12. 11. 01:58 Computer Vision
Michael I. Jordan, Generic constraints on underspecified target trajectories, Proceedings of international conference on neural networks, (1989), 217-225
http://dx.doi.org/10.1109/IJCNN.1989.118584


informed by 함교수님

cf.

Michael I. Jordan



Introduction

connectionist networks

feedforward controller

forward model


activation patterns (in a network)

motor learning

interpretation


visible units & hidden units

The output units of such a network are hidden with respect to learning and yet are visible given their direct connection to the environment.

task space 
articulatory space



Forward models of the environment


The forward modeling approach assumes that the solution to this minimization problem is based on the computation of a gradient.


'Computer Vision' 카테고리의 다른 글

Seam Carving  (0) 2011.01.17
Jordan & Bishop "Neural Networks"  (0) 2010.12.14
Donald M. Wiberg [Schaum's Outline of Theory and Problems of State Space and Linear System]  (0) 2010.12.09
Winsock  (0) 2010.11.26
camera Firefly (FFMV-03M2M)  (0) 2010.11.25
posted by maetel
2010. 12. 9. 22:25 Computer Vision
Schaum's Outline of Theory and Problems of State Space and Linear System (pdf)
Donald M. Wiberg 
McGraw-Hill, 1971 
로욜로 도서관: 2관 4층 515.35 W632s


Preface

"The state space approach is more general than the "classical" Laplace and Fourier transform theory. Consequently, state space theory is applicable to all systems that can be analyzed by integral transforms in time, and is applicable to many systems for which transform theory breaks down"

(1) Linear systems with time-varying parameters can be analyzed in essentially the same manner as time-invariant linear systems.
(2) Problems formulated by state space methods can easily be programmed on a computer.
(3) High-order linear systems can be analyzed.
(4) Multiple input - multiple output systems can be treated almost as easily as single input - single output linear systems.
(5) State space theory is the foundation for further studies such areas as nonlinear systems, stochastic systems, and optimal control.


"Because state space theory describes the time behaviors of physical systems in a mathematical manner, the reader is assumed to have some knowledge of differential equations and of Laplace transform theory."







posted by maetel
2010. 12. 9. 19:00 Footmarks
서강대 수학과 세미나: Neural Network Approximation 
강연: 함남우 교수님 (인천대학교) 

2010-12-09 나무 16:30-17:30 @R1418


함남우 교수님 주전공: 수치해석, 응용수학, 해석학...


> Approximation Theory

- To construct a model for an input/output process
x -> actual process -> f(x)
x -> computed model -> P_f(x)   다항함수

- Error 
Intrinsic Error <= model design
Noise Error <= observation


> Density Problem (접근가능성을 결정)
En(f) ~ inf { || f - p || : p, a subset of Pu }
To decide if En(f) -> 0 as n -> inf
eg. 중간값 정리
n이 얼마나 커야 하는지의 문제


> Complexity Problem
the rate at En(f) -> 0 ( when target fn. f(x) is generally unknown )


* Theory of Best Approximation




ref. G. G. Lorentz, Approximation of Functions


> Brief History of NN
- McCulloch & Pitts, 1943
- Rosenblatt, 1958 "perceptron"
- Minsky & Papert, 1969 "limitation of perceptron"
- Rumelhart, 1986 "hidden layer" (<- 연사님  전공)


> Neural Network (with One Hidden Layer)

x -> Sigma(i=1,...,n) c_i * sigma * (a_i*x + b_i)
a: weight, b: threshold or bias


algebraic polynomial -- approx. --> any continuous fn.


> Complexity Result


* Simultaneous Approximation  (cf. Whitney extension Theorem)

ref. Constructive Function Theory



  


posted by maetel
2010. 12. 1. 22:08

보호되어 있는 글입니다.
내용을 보시려면 비밀번호를 입력하세요.

2010. 11. 26. 23:50 Computer Vision
Winsock informed by jiniseo


by Bob Quinn and Dave Shute, with foreword by Martin Hall
Addison-Wesley




posted by maetel
2010. 11. 25. 13:31 Computer Vision
Firefly (FFMV-03M2M)

http://www.ptgrey.com/products/fireflymv/index.asp

FFMV-03M2M-CS
6p-pin right angle IEEE-1394 Connector
Max 752*480 at 60 FPS
1/3" Micron CMOS , BW
Progressive Scan
Plastic Case Included
FFMV Metal Case
LM5NCL - F1.4 ~1.6 , C-Mount
Adaptor for NMV-4/5WA 렌트 Filter
IR Longpass - 830nm
1394a PCI Adpapter
1394b FWB-LDR-CAT5 Repeater SET


FirePRO LDR
http://www.ptgrey.com/products/firepro/firepro_firewire.asp



Software Requirements (Windows)

  • Windows XP (Service Pack 3 recommended), Windows Vista (Service Pack 1 recommended) or Windows 7
  • MS Visual Studio 6.0 SP5 (to compile and run example code on Windows XP 32-bit operating systems); or
  • MS Visual Studio 2005 SP1 (to compile and run example code on Windows XP 64-bit)
  • MS Visual Studio 2005 SP1 and SP1 Update for Vista (to compile and run example code on Windows Vista or Windows 7)




FlyCapture SKD 2.0
http://www.ptgrey.com/products/pgrflycapture/index.asp

datasheet:


on-line user manual:
http://www.ptgrey.com/support/downloads/documents/flycapture/hh_start.htm

The Point Grey Image Filter Driver (PGRGIGE.sys) was developed for use with GigE Vision cameras. This driver operates as a network service between the camera and the Microsoft built-in UDP stack to filter out GigE vision stream protocol (GVSP) packets.  

The filter driver is installed and enabled by default as part of the FlyCapture SDK installation process. Use of the filter driver is recommended, as it can reduce CPU load and improve image streaming performance.

Point Grey GigE Vision cameras can operate without the filter driver, by communicating directly with the Microsoft UDP stack. GigE Vision cameras operating on Linux systems can communicate directly with native Ubuntu drivers.



FlyCapture SDK 중 이미지 저장 관련 예제 코드 위치:
Program Files > Point Grey Research Inc. > FlyCapture2 > Examples > SaveImageToAviEx
설명: Demonstrates saving a series of images to an AVI file


Cylod (informd by subin)
OpenCV에서 PGR 카메라 사용하는 두가지 방법 소개




OpenCV

Firewire 1394 cameras without Direct X

You can install the CMU 1394 driver for your camera and then use the API of this driver to capture video from the camera. In this way, you can avoid the use of DirectX. See example for details.







posted by maetel