|
Juan Julia LIU Ph. D.
Embedded
Reasoning Area (ERA) Palo
Alto Research Center (PARC) 3333
Coyote Hill Rd, Palo
Alto, CA94304 email: juan.liu@parc.com
telephone: 650-812-4828 |
|
Juan Liu received her B.E. degree in
Electronic Engineering from Tsinghua University,
China, in 1995, and the M.S. and Ph.D. degrees in Electrical Engineering from
the University of Illinois at Urbana-Champaign, in 1998 and 2001, respectively.
In 2001, she joined Palo Alto Research Center (PARC) as a member of research
staff in the Embedded Reasoning Area. Her research interests
include signal processing, statistical modeling and inference, distributed
sensor networks, and applications such as intelligent transportation systems.
She is the recipient of IEEE Signal Processing Society Best Young Author Paper
Award of 2002.
CV and
Publications
RESEARCH
1.
Ad hoc sensor networks and collaborative information
processing
·
Information
processing based on diverse sensing models ---
I have been a key contributor to information processing algorithms for
sensor networks. The application includes detection, classification, and target
tracking. The key method is Bayesian inference.
·
Resource
management in sensor networks --- the challenge in this problem is to achieve a
suitable balance between the need to gather and process a vast amount sensor
data to infer about underlying phenomenon of interest and the reality that
sensor network is often constrained in sensing, processing, and communication
resources. How to achieve best information processing performance without
over-tasking the sensor network is a major challenge. My work along this theme
focuses on developing estimation and information theoretic methods to characterize
the performance of sensor tasking choices, and then using optimization
techniques to decide optimal or nearly optimal solutions.
·
Node
localization in ad hoc networks ---- location information of nodes is critical
to many applications such as routing and sensor tasking. I have designed
localization schemes for a variety of sensing modalities. I have proposed a
unique error characterization and control method to limit error propagation in
sensor network localization.
·
Information
dissemination in ad hoc sensor networks --- designing efficient routing mechanism
for sensor networks, which jointly optimize communication routes balancing the
need for information gathering and delivery.
·
Other
issues ---- programming methodology, middleware, and user interface issues in
sensor networks.
2.
Vehicle-based computation/communication and Intelligent
Transportation System (ITS)
·
Information
dissemination --- designed push-based information dissemination architecture
for VANETs (Vehicle Ad Hoc Networks) by allowing applications to model
potential usefulness of information using utility functions; designed dynamic
packet prioritizing scheme to handle the dynamic and unreliable nature of
VANETs communication. This works better than traditional geocast-like
information dissemination schemes because multiple data sources can trade their
communication resources to maximize overall benefit.
·
Collision
avoidance --- designed computationally efficient statistical inference
mechanism for collision avoidance. The computation is centered at the primary
vehicle, incorporating sensing evidence to monitor all surrounding vehicles and
pedestrians and provide early collision warning.
3.
Statistical modeling
and inference based on acoustic, image, and video signals. The emphasis is mainly on statistical
modeling of various types of signals and applications to estimation problems.
Research interest include: Bayesian estimation, optimization, model selection, wavelets
and multi-resolution signal processing, information theoretic methods, regularization,
and adaptive methods and learning.
4.
Pattern recognition
and computer vision
The focus is on developing computer vision and pattern recognition techniques
for various applications. Example projects include clothes recognition for an
interactive fashion recommender system (PARC project), recognition of natural
cursive handwriting (internship at
Intern Students --- In the past, I had
the privilege to work with these brilliant interns on a number of interesting
topics:
- Dragan Petrovic, UC Berkeley (2002) --- information-directed
routing in sensor networks
- Yi Zou, Duke University (2003)
--- iterative least-square methods for localization
- Arpita Ghosh, Stanford University (2004) --- variable
information dissemination
- Manu Chhabra,
University of Rochester (2004) --- resource management in monitoring
cluttered environment with limited sensing resources
- Xiaotao Zou, UC Santa
Barbara (2005) --- camera network
calibration
- Deepan Thavarajah (2007) ---
collision early warning system implementation
- Wei Zhang, Oregon State
University (2007) --- responsive mirror for intelligent fitting room using
multi-camera perception
AWARDS AND
PROFESSIONAL ACTIVITIES
·
Member
of IEEE and ACM.
·
Recipient of IEEE Signal Processing Society
Young Author Best Paper Award (2002) for the paper "Information-Theoretic
Analysis of Interscale and Intrascale
Dependencies Between Image Wavelet Coefficients",
coauthored with Pierre Moulin.
·
Guest
editor of IEEE Signal Processing Magazine, special issue on distributed signal
processing in sensor networks.
·
Finance
Chair, IPSN 2003.
·
Technical
Program Committee (TPC) member of various workshops and conferences, including
IPSN 2004 and 2006 (Workshop on Information
Processing in Sensor Networks), SensorFusion
2005, ICIP (International Conference on Image Processing) 2003 and 2004.
·
Reviewer
for various journal and magazine, including IEEE Transactions on Image
Processing, IEEE Transactions on Signal Processing, IEEE Signal Processing Megazine, IEEE Pattern Recognition Letter, ACM Transactions
on Sensor networks, EuraSip Journal on Applied Signal
Processing, Signal Processing of Elseviser
Science.