Miao Liao

miao DOT liao AT uky DOT edu

Gravity Lab (Graphics Visions Technology Laboratory)

Center for Visualization and Virtual Environments

Department of Computer Science, University of Kentucky

Advisor: Dr. Ruigang Yang

 

I started my Ph.D. study in Fall 2005 at the Department of Computer Science, University of Kentucky (UKY). My research interests lie in Computer Vision, Image Processing, Computer Graphics and Multimedia. My advisor is Dr. Ruigang Yang. Before I came to UKY, I received my Bachelor's degree in School of Software (2005), Tsinghua University, P. R. China. I will work as a Senior Display Algorithm Researcher in Sharp Laboratories of America soon.

 

Curriculum Vitae (PDF Version)


EXPERIENCE

Teaching Assistant, Computesr Science Department, University of Kentucky, Spring 2011

Research Assistant, Computesr Science Department, University of Kentucky, with Dr. Ruigang Yang, 8/2005~1/2011

Intern, Intuitive Surgical Inc., Sunnyvale CA, with Dr. Wenyi Zhao, 8/2008~10/2008

Intern, Microsoft Research, Redmond WA, with Dr. Zhengyou Zhang and John Lewis, 5/2007~8/2007

Intern, Microsoft Research, Redmond WA, with Dr. Zhengyou Zhang, 1/2007

Student Research Training, CG&CAD Center, Tsinghua University, 8/2004~2/2005

Summer Research Intern, CG&CAD Center, Tsinghua University, 5/2004~8/2004


PUBLICATIONS

Interreflections Removal for Photometric Stereo by Using Spectrum-dependent Albedo (pdf)
Miao Liao, Xinyu Huang, Ruigang Yang, to appear in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011.

Video Stereolization: Combining Motion Analysis with User Interaction (pdf)
Miao Liao, Jizhou Gao, Ruigang Yang, Minglun Gong, accepted to IEEE Transactions on Visualization and Computer Graphics (TVCG).

Complete 3D Model Reconstruction Using Two Types of Depth Sensors(pdf)
Guangyu Mu, Miao Liao, Ruigang Yang, Dantong Ouyang, Zhiwen Xu, Xiaoxin Guo,. IEEE International Conference on Intelligent Computing and Intelligent Systems , 2010.

Complete 3D Model Reconstruction Using a Depth Sensor(pdf)
Guangyu Mu, Miao Liao, Ruigang Yang, Dantong Ouyang, Zhiwen Xu, Xiaoxin Guo,. IEEE International Conference on Intelligent Computing and Integrated Systems , 2010.

A Volumetric Approach for Merging Range Images of Semi-Rigid Objects Captured at Different Time Instances(pdf)
Miao Liao, Qing Zhang, Ruigang Yang and Minglun Gong. 3DPVT, 2010

Modeling Deformable Objects from A Single Depth Camera (pdf)
Miao Liao, Qing Zhang, Huamin Wang , Ruigang Yang and Minglun Gong. ICCV, 2009 (Oral Presentation)

Physically Guided Liquid Surface Modeling from Videos (pdf)
Huamin Wang, Miao Liao, Qing Zhang, Ruigang Yang and Greg Turk. SIGGRAPH, 2009

Joint Depth and Alpha Matte Optimization via Fusion of Stereo and Time-of-Flight Sensor (pdf)
Jiejie Zhu, Miao Liao, Ruigang Yang and Zhigeng Pan. CVPR, 2009

Software-based Distortion Comensation for a Scanned Beam Display (pdf)
Miao Liao, Zhengyou Zhang and John Lewis. IEEE International Workshop on Projector-Camera Systems (ProCams), 2008

Real-Time Light Fall-off Stereo (pdf)
Miao Liao, Liang Wang, Ruigang Yang and Minglun Gong. IEEE International Conference on Image Processing (ICIP), 2008

Light Fall-off Stereo (pdf)
Miao Liao, Liang Wang, Ruigang Yang and Minglun Gong. IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), 2007

Robust and Accurate Visual Echo Cancellation in a Full-duplex Projector-camera System (pdf)
Miao Liao, Ruigang Yang and Zhengyou Zhang. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), September 2007

Robust and Accurate Visual Echo Cancellation in a Full-duplex Projector-camera System (pdf)
Miao Liao, Mingxuan Sun, Ruigang Yang and Zhengyou Zhang. IEEE International Workshop on Projector-Camera Systems (ProCams), New York, NY, USA, June 17, 2006

Real-time Global Stereo Matching Using Hierarchical Belief Propagation (pdf)
Qingxiong Yang, Liang Wang, Ruigang Yang, Shengnan Wang, Miao Liao and David Nister. The British Machine Vision Conference (BMVC), 2006

High Quality Real-time Stereo using Adaptive Cost Aggregation and Dynamic Programming (pdf)
Liang Wang, Miao Liao, Minglun Gong, Ruigang Yang and David Nister. Third International Symposium on 3D Data Processing, Visualization and Transmission (3DPVT), 2006 (Oral Presentation)

PAPERS UNDER REVIEW

Towards Real-time Joint Depth and Alpha Matte Optimization via Fusion of Stereo and Time-of-Flight Sensor
Miao Liao, Jiejie Zhu, Ruigang Yang and Zhigeng Pan. Submitted to IAPR Machine Vision and Application (MVA).

Light Fall-off Stereo
Miao Liao, Liang Wang, Ruigang Yang and Minglun Gong. Submitted to IEEE Transaction on Image Processing (TIP).

 


PROJECTS

Video Stereolization

We present a semi-automatic system that converts conventional videos into stereoscopic videos by combining motion analysis with user interaction, aiming to transfer as much as possible labeling work from the user to the computer. In addition to the widely-used structure from motion (SFM) techniques, we develop two new methods that analyze the optical flow to provide additional qualitative depth constraints. They remove the camera movement restriction imposed by SFM so that general motions can be used in scene depth estimation ¨C the central problem in mono-to-stereo conversion. With these algorithms, the user¡¯s labeling task is significantly simplified. We further developed a quadratic programming approach to incorporate both quantitative depth and qualitative depth (such as these from user scribbling) to recover dense depth maps for all frames, from which stereoscopic view can be synthesized. In addition to visual results, we present user study results showing that our approach is more intuitive and less labor intensive, while producing 3D effect comparable to that from current state-of-the-art interactive algorithms.

Single View Deformable Object Modeling

We propose a novel approach to reconstruct complete 3D deformable models over time by a single depth camera, provided that most parts of the models are observed by the camera at least once. The core of this algorithm is based on the assumption that the deformation is continuous and predictable in a short temporal interval. While the camera can only capture part of a whole surface at any time instant, partial surfaces reconstructed from different times are assembled together to form a complete 3D surface for each time instant, even when the shape is under severe deformation. A mesh warping algorithm based on linear mesh deformation is used to align different partial surfaces. A volumetric method is then used to combine partial surfaces, fix missing holes, and smooth alignment errors. Our experiment shows that this approach is able to reconstruct visually plausible 3D surface deformation results with a single camera. Related: paper, video

Liquid Modeling from Videos

We present an image-based reconstruction framework to model real water scenes captured by stereoscopic video. In contrast to many image-based modeling techniques that rely on user interaction to obtain high-quality 3D models, we instead apply automatically calculated physically-based constraints to refine the initial model. The combination of image-based reconstruction with physically-based simulation allows us to model complex and dynamic objects such as fluid. Using a depth map sequence as initial conditions, we use a physically based approach that automatically fills in missing regions, removes outliers, and refines the geometric shape so that the final 3D model is consistent to both the input video data and the laws of physics. Physically-guided modeling also makes interpolation or extrapolation in the space-time domain possible, and even allows the fusion of depth maps that were taken at different times or viewpoints. We demonstrated the effectiveness of our framework with a number of real scenes, all captured using only a single pair of cameras. Related: paper, video

Matting Via Depth

We present a new approach to iteratively estimate both high-quality depth map and alpha matte from a single image or a video sequence. Scene depth, which is invariant to illumination changes, color similarity and motion ambiguity, provides a natural and robust cue for foreground/background segmentation ¨C a prerequisite for matting. The image mattes, on the other hand, encode rich information near boundaries where either passive or active sensing method performs poorly. We develop a method to combine the complementary nature of scene depth and alpha matte to mutually enhance their qualities. We formulate depth inference as a global optimization problem where information from passive stereo, active range sensor and matte is merged. The depth map is used in turn to enhance the matting. In addition, we extend this approach to video matting by incorporating temporal coherence, which reduces flickering in the composite video. We show that these techniques lead to improved accuracy and robustness for both static and dynamic scenes. Related: paper, code, video

Laser Projector Simulator

We developed a software based approach to compensate for distortion in a scanned beam display (SBD) system. A scanned beam display suffers from two main problems: the non-uniform power spreading within a single light spot and the non-uniform light spot distribution in trajectory. While most work has been done to solve these problems from the hardware design perspective, we address these issues by minimizing the difference between input and output images on a software level. We also implemented our algorithm on Graphics Processing Unit (GPU) to achieve near real-time performance which is required by video display. Finally, the output light source control value is applied to a scanned beam display simulator, showing improved display images. Related: paper

Light Fall-off Stereo

We present a real-time depth recovery system using Light Fall-off Stereo (LFS). Our system contains two co-axial point light sources (LEDs) synchronized with a video camera. The video camera captures the scene under these two LEDs in complementary states(e.g., one on, one off). Based on the inverse square law for light intensity, the depth can be directly solved using the pixel ratio from two consecutive frames. We demonstrate the effectiveness of our approach with a number of real world scenes. Quantitative evaluation shows that our system compares favorably to other commercial real-time 3D range sensors, particularly in textured areas. We believe our system offers a low-cost high-resolution alternative for depth sensing under controlled lighting. Relate: CVPR07, ICIP08 , video

Visual Echo Cancellation

We developed a comprehensive set of techniques to address the ¡°visual echo¡± problem in a full-duplex projectorcamera system. A calibration procedure records the geometric and photometric transfer between the projector and camera in a look-up table. With the calibration information, the predicted camera view of the projected image is compared against the captured camera image to find echo pixels. Only non-echo pixels are used for display, therefore achieving the goal of suppressing visual echo. Compared to previous techniques, our approach¡¯s main advantages are two-fold. First, it accurately handles full color images with no assumption about the surface reflectance or the photometric response of the projector or camera. Secondly, it is robust to geometric registration errors and quantization effect. It is particularly effective for high-frequency contents such as texts and hand drawings. We demonstrate the effectiveness of our approach with a variety of real images in a full-duplex projector-camera system. Related: Procams06, TPAMI07, video

Real-time Stereo Matching

We present a stereo algorithm that achieves high quality results while maintaining real-time performance. The key idea is simple: we introduce an adaptive aggregation step in a dynamic-programming (DP) stereo framework. The per-pixel matching cost is aggregated in the vertical direction only. Compared to traditional DP, our approach reduces the typical ¡°streaking¡± artifacts without the penalty of blurry object boundaries. Evaluation using the benchmark Middlebury stereo database shows that our approach is among the best (ranked first in the new evaluation system) for DP-based approaches. The performance gain mainly comes from a computationally expensive weighting scheme based on color and distance proximity. We utilize the vector processing capability and parallelism in commodity graphics hardware to speed up this process over two orders of magnitude. Over 50 million disparity evaluations per second (MDE/s)1 are achieved in our current implementation.Related: 3DPVT06, BMVC06, video


AWARDS AND GRANTS

Kentucky Opportunity Fellowship, University of Kentucky, 2010~2011.
IEEE PAMI-Technical Committee travel grants, ICCV, 2009.
Conference and Research Student Support Funding, University of Kentucky, 2007
Conference and Research Student Support Funding, University of Kentucky, 2006
Full Research Assistantship, University of Kentucky, 2005~present.


TECHNICAL TALKS

Paper presenation on 3DPVT 2010, Paris, France
Paper presentation on ICCV 2009, Kyoto, Japan
Final internship reporting in Intuitive Surgical Inc., CA 2008
Paper presentation on ProCams 2008, Los Angeles, CA
Paper presentation on Mid-West Graphics workshop 2007, University of Iowa, Iowa City, IA
Final intern reporting in Microsoft Research, Redmond, WA, 2007
Paper presentation on Mid-West Graphics workshop 2006, Vanderbilt University, Nashville, TN
Paper presentation on ProCams 2006, New York City


PROFESSIONAL SERVICE

Program committee member of International Conference on Computer Vision (ICCV) 2011
Invited Reviewer of Graphics Interface (GI) 2011
Invited Reviewer of Pattern Recognition Letters
Invited Reviewer of IEEE Computer Vision and Pattern Recognition (CVPR)2011
Invited Reviewer of IEEE International Journal of Computer Vision (IJCV)
Invited Reviewer of IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
Invited Reviewer of IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)
Invited Reviewer of EURASIP Journal on Image and Video Processing
Invited Reviewer of International Journal of Imaging
Invited Reviewer of IET Computer Vision
Invited Reviewer of MDPI Sensors
Invited Reviewer of IAPR Machine Vision and Application (MVA)


ACTIVITIES AND MEMBERSHIP

Volunteer in the security team of the 2010 Alltech FEI World Equestrian Games
President of Chinese Student & Scholar Association, University of Kentucky, 2007~2008
Member of IEEE
Member of ACM
Member of Sigma Xi
Member of Microsoft Research Alumni


INVENTION DISCLOSURE

Light Fall-off Stereo: Invention disclosure at the University of Kentucky, 2006

Fast and High Accuracy Tooth Reconstruction Sensor : Invention disclosure at the University of Kentucky, 2011


HOBBY

I take photography very seriously and I am marketing my photos on Shutterstock in free time. I belong to the .33 club of Shutterstock. Please visit my image gallery through this link http://shutterstock.com/g/liaomiao. If you plan to sell your photos on Shutterstock after seeing this, please indicate me as referrer by visiting the Shutterstock Submit site via this link: http://submit.shutterstock.com/?ref=321835, and my referrer code is 321835. If you plan to purchase any photos on Shutterstock, please also point out me as referrer by visit Shutterstock via this link: http://www.shutterstock.com/?rid=321835.

My latest images for sale at Shutterstock:

My most popular images for sale at Shutterstock:

 

I am also rendering pictures that is hard to be captured by cameras. I use 3ds Max + Vray Ray Tracer. The photography and 3D rendering are actually moving towards each other. Photography makes the best use of the camera, lens, lighting and post-processing to generate deficiency-free photos that look like 3D rendering. While 3D rendering tries to simulate the defect of the real world to make its output like a photo. Can you tell which one is photo and which one is rendering in the following two pictures:

Copyright of Fotoopa Copyright of Gilles Tran

 

 


 

CONTACTS

Miao Liao

Center for Visualization and Virtual Environments
University of Kentucky
1 Quality Street, Suite 800
Lexington, KY 40507-1464, USA

Phone: +1-859-257-1257 ext 82332
E-mail: miao DOT liao AT uky DOT edu

Homepage: http://vis.uky.edu/~liaomiao/

Web Site Hit Counters
Web Counters

  


Last Modified on Monday, May 20, 2008.