發信人: GzLi (笑梨), 信區: DataMining
標  題: 特征選擇講座主講介紹
發信站: 南京大學小百合站 (Sun Dec  1 23:55:25 2002), 站內信件


劉燕西：http://www-2.cs.cmu.edu/~yanxi/home.html
Research Scientist in CMU
Associated centers: VASC and MRTC
Email address: yanxi@cs.cmu.edu
Mailing address:
Carnegie Mellon University
Robotics Institute
5000 Forbes Avenue
Pittsburgh, PA 15213 

Dr. Liu's research focus is on learning semantically discriminative image fea
tures from large biomedical image datasets. Her computational tools are drawn
 from statistical learning theory, information theory,
pattern recognition, image processing and computer vision. The goal of her re
search is to seek the hidden patterns in large image databases, by revealing 
their intrinsic dimensionality and separability. 

Research interests 


My research interests span a range of applications in computer vision and rob
otics, with a central research theme: computational symmetry. Computational s
ymmetry addresses issues on robust representation, detection, and reasoning a
bout symmetry, as well 
as diverse applications of applying (a)symmetry analysis on computers (see pr
ojects). 

Symmetry is a pervasive phenomena in both natural and man-made environments. 
Humans have an innate ability to perceive and take advantage of symmetry in e
veryday life, but it is not obvious how to automate this powerful insight on 
man-made intelligent 
beings, such as robots. On the surface, symmetry is simple and basic. In esse
nce, the concept of symmetry is much more than a mirror reflection with binar
y choices, rather, it can span a continuous spectrum of multi-dimensional spa
ces. 

In basic sciences, the understanding of symmetry played a profound role in se
veral important discoveries including: relativity theory (the symmetry of tim
e and space); human DNA structure (double helix); the quasicrystals and their
 mathematical 
counterpart penrose tiles. We argue that reasoning about symmetry can likewis
e play a crucial part in the advance of artificial/machine intelligence. 

A computational model for symmetry is especially pertinent to robotics, compu
ter vision and machine intelligence, because in these fields we are studying 
how a man-made intelligent being can perceive and interact with the chaotic r
eal world in the most 
effective way. Recognition of symmetries is the first step towards capturing 
the essential structure of a real world problem, and minimizing redundancy wh
ich can often lead to drastic reductions in computation. One fundamental limi
tation of computers 
is their finite representational power. One simple floating point error can d
estroy any perfect symmetry. One's ability to tolerate departure from perfect
 symmetry reflects one's level of sophistication in perception, which need to
 be built into the 
development of machine/artificial intelligence. Besides our poor understandin
g of human?s natural capability of symmetry perception, the mathematical theo
ry for symmetry, group theory, has not been utilized effectively in practice.
 Group theory is 
usually introduced in classrooms in an abstract way (if it is introduced to c
omputer science majors at all in the United States) that is hard to relate to
 everyday life. More importantly, the non-coherent topological nature of symm
etry groups poses 
challenging computation problems on computers. I am finishing up a textbook f
or engineering students to learn group theory through concrete examples from 
applications in robotics (assembly planning) and computer vision (repeated pa
ttern perception). 

My current projects related to computational symmetry include: 

1) Brain Asymmetry 


using quantitative, statistical image features extracted from 3D volumetric r
adiology neuroimages (MR, CT, PET ...) to find similar brains with the same p
athology; 
using asymmetry measure of MRI human brains to detect schizophrenia patients; 
from large, population-based image databases, we are also interested in findi
ng out the answer to this question: 
how symmetrical are the normal (human, mice, . . . ) brains? 

2) Facial Asymmetry 

Using facial asymmetry as a biometric to identify faces under expression, pos
t and lighting variations. The questions we are seeking the answers for are: 


is facial asymmetry a characteristic of human identity or expression? 
does facial asymmetry remain relatively invariant under expression variations? 
are more attractive people more recognizable? 
how facial asymmetry changes within the invisible wavelengths (infrared, ther
mal)?
3) Repeated Pattern Perception using Crystallographic Groups 

What do you see when you look at a regularly textured surface? do you see til
es? or do you see structures? We are developing a computational model for rep
eated pattern perception that is able to automatically classify a given patte
rn into one of the 7 
frieze groups (patterns repeating along one direction), or one of the 17 wall
paper groups (patterns repeating along two linearly independent directions), 
or one of the 230 space groups (patterns repeating in 3D Euclidean space). It
 can also 
automatically generate a finite set of possible tiles (based on our theoretic
al proofs). Furthermore, we study repeated patterns under different viewing d
irections to find out what happens to a periodic pattern when it is deformed 
by Affine or 
perspective transformations? 

4) Gait analysis using wallpaper groups 

Spatiotemporal representation of human or animal gaits form a naturally appre
ciable periodic pattern. Different gaits are reflected by different symmetrie
s and symmetry groups of such patterns. We study the possibility of using cue
s extracted from such 
patterns for identity and activity classification. 

In addition to computational symmetry, I am interested in discovering hidden 
patterns from large image sets, in particular, large biomedical image databas
es. These images are especially attractive for studying image meanings since 
they are normally 
associated with unambiguous, objective underlying semantics from physical cau
ses (versus images that can be interpreted one way or the other subjectively 
by the viewer). With the worldwide trend towards ?paperless? hospitals, the c
ommercially available 
Picture Archiving and Communication System (PACS) installed in many hospitals
 collects a large amount of digital biomedical image data monthly, weekly eve
n daily. However, the utilization of such data for research and education is 
hampered by the lack 
of intelligent, effective image analysis, comparison and retrieval tools. 

My research focus is to learn semantically discriminative image features usin
g statistical learning theory, information theory, and pattern recognition, i
mage processing and computer vision tools. The goal is to seek the true funda
mental dimensionality 
and separability of a given image set and image features. The philosophy of o
ur approach is "un-biased least commitment", and it is executed as follows: 


image features are extracted extensively and creatively; 
image features are selected objectively; 
no image feature is excluded without strong quantitative justifications. 
We close the learning loop from imaging process --> image feature extraction 
--> image feature screening --> image feature grouping --> image feature subs
et selection --> image classification and image retrieval. We have applied th
ese ideas in multiple 
application domains (pathological neuroimages, facial expression videos, mult
ispectral microscopic images) with very promising results (see our publicatio
ns). We have several on-going projects exploring along these research directi
ons intensively (see 
our projects). The results from this research are directly applicable to the 
fast growing biomedical informatics industry and hospitals, with which we hav
e and we continue establishing tight collaborations.

 --               ***  端莊厚重 謙卑含容 
事有歸著 心存濟物  *** 數據挖掘  http://DataMining@bbs.nju.edu.cn/ 

※ 來源:．南京大學小百合站 bbs.nju.edu.cn．[FROM: 211.80.38.17]