CAREER: High-Order Tensor Analysis for Groupwise Correspondence: Theory, Algorithms, and Applications


Investigator:
Haibin Ling, PhD
Center for Data Analytics and Biomedical Informatics
Dept. of Computer & Information Sciences
305 Wachman Hall, Temple University
1805 N. Broad St., Philadelphia, PA 19122
(tel)1-215-204-6973 | (fax)1-215-204-5082
(email) hbling AT temple.edu

Supported by:
     
NSF project link
      

Summary:
Visual matching is a fundamental problem in computer vision (CV) and intensive research efforts have been devoted to building correspondence between a pair of visual objects. By contrast, finding correspondence among an ensemble of objects remains challenging. This project develops a unified framework for this problem and to apply the framework to different applications. The research establishes a close correlation between the classical multi-dimensional assignment (MDA) problem and low-rank tensor approximation. Such correlation paves a way of using high-order tensor analysis for groupwise visual matching that assumes an MDA formulation. Along the way, a series of algorithms are developed to address challenging issues such as computational efficiency and context modeling. These algorithms are then deployed to different tasks including simultaneous tracking of multiple targets, tracking of deformable structures, and batch alignment of visual ensembles.

This project can generate broad impact on areas of computer vision, computer graphics, combinatorial optimization, oral and maxillofacial radiology, image-guided intervention, physical therapy, security and defense, education research, etc. On the one hand, the fundamental importance of visual matching makes the project transformative to many other CV problems. On the other hand, the project benefits a wide range of fields outside the CV community through the use of interdisciplinary applications as test beds. This project also integrates tightly research and education with highlights on supervising students from underrepresented groups, combining computer vision and education research, and involving undergraduates in research.

Objectives:
The aims of the project include (1) a general framework of using high-order tensor analysis for groupwise correspondence, (2) applying the framework to computer vision tasks including multi-target tracking, deformable tracking, and batch alignment, (3) integrating the developed algorithms to interdisciplinary research areas such as medical image analysis and aerial video surveillance, and (4) deploying developed algorithms for the research community.

Publications (links to data and code provided whenever applicable)

Osteoporosis Prescreening Using Dental Panoramic Radiographs Feature Analysis
C. Bo, X. Liang, P. Chu, J. Xu, X. Wang, J. Yang, V. Megalooikonomou, and H. Ling
Proc. of IEEE Int'l Symposium on Biomedical Imaging (ISBI), 2017
PDF
Branching Path Following for Graph Matching
T. Wang, H. Ling, C. Lang, and J. Wu
In Proc. of European Conference on Computer Vision (ECCV), 2016.
PDF | code
Structure-Aware Rank-1 Tensor Approximation for Curvilinear Structure Tracking Using Learned Hierarchical Features
P. Chu, Y. Pang, E. Cheng, Y. Zhu, Y. Zheng, and H. Ling
International Conf. on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2016.
PDF
Symmetry-Aware Graph Matching
T. Wang, H. Ling, C. Lang, and H. Yang
Pattern Recognition (PR), 60:657-668, 2016
Salient Object Detection via Structured Matrix Decomposition
H. Peng, B. Li, H. Ling, W. Hu, W. Xiong, and S. Maybank
IEEE Trans. on Pattern Analysis and Machine Intelligence (PAMI), 2016.
code | data | benchmark |
Light Mixture Intrinsic Image Decomposition Based on a Single RGB-D Image
G. Xing, Y. Liu, W. Zhang, and H. Ling
The Visual Computer, 32(6-8): 1013-1023, 2016
PDF
Tensor Power Iteration for Multi-Graph Matching
X. Shi, H. Ling, W. Hu, J. Xing, and Y. Zhang
IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2016.
PDF |
Path Following with Adaptive Path Estimation for Graph Matching
T. Wang and H. Ling
Proc. of AAAI Conference on Artificial Intelligence (AAAI), 2016
PDF
Dynamic Scene Classification Using Redundant Spatial Pooling
L. Du and H. Ling
IEEE Trans. on Cybernetic, 46(9):2156-2165, 2016
PDF
3D Hand Pose Estimation Using Randomized Decision Forest with Segmentation Index Points
P. Li, H. Ling, X. Li, and C. Liao
IEEE International Conference on Computer Vision (ICCV), 2015.
PDF
Encoding Color Information for Visual Tracking:Algorithms and Benchmark
P. Liang, E. Blasch, and H. Ling
IEEE Trans. on Image Processing (T-IP), in press
PDF | Download Benchmark
Predicting Image Memorability by Multi-view Adaptive Regression
H. Peng, K. Li, B. Li, H. Ling, W. Xiong, and W. Hu
ACM Multimedia Conference (ACM MM), 2015.
PDF
Cross-Age Face Verification by Coordinating with Cross-Face Age Verification
L. Du and H. Ling
IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2015.
PDF | Source code
Covert Photo Classification by Fusing Image Features and Visual Attributes
H. Lang and H. Ling
IEEE Trans. on Image Processing (T-IP), 24(10):2996--3008, 2015
dataset
Robust Nighttime Vehicle Detection by Tracking and Grouping Headlights
Q. Zou, H. Ling, S. Luo, Y. Huang, and M. Tian
IEEE Trans. on Intelligent Transportation Systems (T-ITS), in press
GARP-Face: Balancing Privacy Protection and Utility Preservation in Face De-identification
L. Du, M. Yi, E. Blasch, and H. Ling
Proc. of IEEE Int'l Joint Conf. on Biometrics (IJCB), Clearwater, 2014.
PDF |
Exploiting Competition Relationship for Robust Visual Recognition
L. Du and H. Ling
Proc. of AAAI Conference on Artificial Intelligence (AAAI), 2014
PDF | Code
Bin Ratio-Based Histogram Distances and Their Application to Image Classification
W. Hu, N. Xie, R. Hu, H. Ling, Q. Chen, S. Yan, and S. Maybank
IEEE Trans. on Pattern Analysis and Machine Intelligence (PAMI), 36(12): 2338-2352, 2014.
PDF |
Transfer Learning Based Visual Tracking with Gaussian Process Regression
J. Gao, H. Ling, W. Hu, and J. Xing
In Proc. of European Conference on Computer Vision (ECCV), Part III, LNC 8691, pp. 188-203, 2014.
PDF | Supplementary | Code |
Discriminative Vessel Segmentation in Retinal Images by Fusing Context-Aware Hybrid Features
E. Cheng, L. Du, Y. Wu, Y. Zhu, V. Megalooikonomou, and H. Ling
Machine Vision and Applications (MVA), 25(7): 1779-1792, 2014.
PDF
Multi-target Tracking with Motion Context in Tenor Power Iteration
X. Shi, H. Ling, W. Hu, C. Yuan, and J. Xing
IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Columbus OH, 2014
PDF
Curvilinear Structure Tracking by Low Rank Tensor Approximation with Model Propagation
E. Cheng, Y. Pang, Y. Zhu, J. Yu, and H. Ling
IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Columbus OH, 2014
PDF