Assistant Professor
Center for Information Science & Technology
Dept. of Computer & Information Sciences
324 Wachman Hall, Temple University
1805 N. Broad St., Philadelphia, PA 19122
(tel)1-215-204-6973 | (fax) 1-215-204-5082 | hbling AT temple.edu
| |
Haibin Ling Assistant Professor Center for Information Science & Technology Dept. of Computer & Information Sciences 324 Wachman Hall, Temple University 1805 N. Broad St., Philadelphia, PA 19122 (tel)1-215-204-6973 | (fax) 1-215-204-5082 | hbling AT temple.edu |
Short Bio:
Haibin Ling received the B.S. degree in mathematics and the MS degree in computer
science from Peking University,
China, in 1997 and 2000, respectively, and the PhD degree from the
University of Maryland, College Park,
in Computer Science in 2006.
From 2000 to 2001, he was an assistant researcher in the Multi-Model User Interface Group
at Microsoft Research Asia.
From 2006 to 2007, he worked as a postdoctoral scientist at the
University of California Los Angeles.
After that, he joined Siemens Corporate Research
as a research scientist. Since fall 2008, he has been an Assistant Professor at
Temple University.
Dr. Ling's research interests include computer vision, medical image analysis,
human computer interaction, and machine learning. He received the Best Student
Paper Award at the ACM Symposium on User Interface Software and Technology (UIST)
in 2003.
|
We present a hierarchical, learning-based approach for automatic liver segmentation from 3D CT volumes. We target volumes coming from largely diverse sources and generated by different scanning protocols. Three key ingredients are combined to solve the problem: a hierarchical framework, learning techniques, and shape space initialization. The proposed approach is tested on a challenging dataset containing 174 volumes. It not only produces excellent segmentation accuracy, but also runs about fifty times faster than state-of-the-art solutions. |
Hierarchical, Learning-based Automatic Liver Segmentation, CVPR'08.
|
We propose the proximity distribution kernels (PDK) to simulateneously address the spatial relation between local features, while be robust to geometric transformations such as translation, rotation, and scaling. PDK forms a Mercer kernel and is readily combined with kernel machines such as SVM. We tested it for category classification tasks on three public datasets, Graz-I, Graz-II, and PASCAL Challenge 05. The PDK based approaches outperformed all previously reported methods. |
Proximity Distribution Kernels for Geometric Context in Category Recognition, ICCV'07.
|
Face recognition across ages is an important but challenging task due to the large image variation caused over time. We propose using the gradient orientation pyramid for this task. Discarding the gradient magnitude and utilizing hierarchical techniques, we find that the new descriptor yields a robust and discriminative representation. In addition, our experiments show that, although the aging process adds difficulty to the recognition task, it does not surpass illumination or expression as a confounding factor. |
A Study of Face Recognition as People Age, ICCV'07.
|
Diffusion Distance. We model the difference between histograms as a temperature field and
study the relationship between histogram similarity and a diffusion process,
showing how diffusion handles deformation as well as quantization
effects. As a result, the diffusion distance is derived as the sum of dissimilarities
over scales, which is robust to deformation, lighting change and noise. In
addition, it enjoys linear computational complexity which significantly improves
previously proposed cross-bin distances with quadratic complexity or higher. Diffusion Distance for Histogram Comparison, CVPR'06.
|
|
EMD-L1. We propose a fast algorithm, EMD-$L_1$, for computing the Earth Mover's Distance (EMD)
between histograms. Compared to the original formulation, EMD-$L_1$ has a
largely simplified structure and is equivalent to the original EMD with
L_1 ground distance. Exploiting the L_1 metric structure, an
efficient tree-based algorithm is designed to solve the EMD-$L_1$ computation. An
empirical study shows that EMD-L1 has a $O(N^2)$ complexity,
which is much faster than previously reported algorithms with super-cubic complexities.
An Efficient Earth Mover's Distance Algorithm for Robust Histogram Comparison, PAMI'07. |
|
Deformation Invariant Descriptor. We propose a novel framework to build descriptors of local intensity that are
invariant to general deformation. In this framework, an image is embedded as a 2D
surface in 3D space, with intensity weighted relative to distance in $x$-$y$. We show
that as this weight increases, geodesic distances on the embedded surface are less
affected by image deformations. In the limit, distances are deformation invariant. We use
geodesic sampling to get neighborhood samples for interest points, then use a
geodesic-intensity histogram (GIH) as a deformation invariant local descriptor.
Deformation Invariant Image Matching, ICCV'05.
ICCV Talk (4.5M).
|
|
Inner-Distance. We found that the inner-distance which is defined as the shortest path length in shapes
is (invariant) insensitive to (ideal) articulations. We demonstrated that it
can be used to build very discriminative descriptors for shapes with parts,
especially with articulations. In the experiments on several widely tested
dataset including the MPEG7 shape dataset and the Kimia dataset, our approach
outperforms all other reported methods.
Shape Classification Using the Inner-Distance, PAMI'07.
Articulated shape data
|
First Steps Toward an Electronic Field Guide for Plants, Taxon'06. Searching the World's Herbaria: A System of Visual Identification of Plant Species, ECCV'08. 
|
Use machine learning techniques, we designed methods for the specimen images processing, including frame removing and background elimination and leaf/stem labeling. The basic idea is to combine k-means and SVM on region-based features. Currently we are doing further experiments on leaf-segmentation using graph models and tryiing to apply the statistical shape theory to capture the leaf deformations. |
 
|
This part of work focuses on leaf image classification. We designed experiments on some leaf image databases, including Fourier descriptor, shape context, and inner-distance shape context. Beside of that, some early experiments for model based leaf detection from specimen image are conducted, where we combine the mean-shift and the generalized Hough transform. We use the statistical method to model the leaf shapes to allow deformations. A short report of the earlier work can be found at report at Smithsonian, 09/2003 (6M) To show that the detection method is generic, we also tested the method with several medical images from Falzenswalb's paper. |
|
We are trying to find new approaches for shape representation, which 1) are
invairant to some geometric transforms, 2) allow natually combination of the
texture and the shape information, and 3) are as informative as possible.
In the experiments, we studied the scale invariant and affine invariant features (Lowe, Mikolajczyk and Schmid). We also tested the inner-distance with the MDS to simulate the 2D version of Elad and Kimmel's bending invariant works. Furthermore, we tried to design a new features for the goal mentioned above. |
|
When images are shrunk into thumbnails, they often become difficult to browse.
We attack this problem by cropping the informative part of an image before
shrinking it. The informativeness is measured by the saliency map which is
widely studied in the vision area. The algorithm is initially completed as
an independent study, then we systematically did experiments to show the effectiveness
of the algorithms. Details can be found in my independent study or our UIST 2003 paper. Independent Study Experiment,
Project Report Automatic Thumbnail Cropping and Its Effectiveness, ACM UIST 2003.
|