Selecting Objects with Freehand Sketches

Selection in digital image editing is the task of extracting an object embedded in an image, a task performed frequently in many digital content creation applications. Available tools either have severe limitations in their capabilities, or require very careful user guidance and control. In this project, we attempt to making selection easier and usable in a wider range of applications. We present the design of a tool for selecting objects using simple, rough freehand sketches similar to those used in normal interhuman communication. more

K.-H. Tan and N. Ahuja. A Representation for Image Structure and Its Application in Object Selection with Freehand Sketches. In proceedings IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR) 2001. PS 13.2MB PDF 4.3MB Small PDF 1.3MB

K.-H. Tan and N. Ahuja. Selecting Objects with Freehand Sketches. In proceedings IEEE International Conference on Computer Vision (ICCV) 2001.IEEE Xplore Citation PS 19.8MB PDF 12.9MB Small PDF 1.3MB

Multi-flash Imaging

Distinguishing 3D outlines from texture edges is one of the long standing challenges in computer vision and graphics. We show that a consumer digital camera augmented with multiple flashes can reliably recover depth edges, or discontinuities in depth. The camera can also remove shadows to a large extent and the technique can be implemented even on medical laparoscopes with fiber optic illumination.

Photo.Net Slashdot

SIGGRAPH Highlighted Paper Highlighted Emerging Technology activity (More) Electronic Theater (conference video highlight)

Formation Control using Virtual Structures

A key problem in cooperative robotics is the maintenance of a geometric configuration during movement. To address this problem, the concept of a virtual structure is introduced. Control methods are developed to force an ensemble of robots to behave as if they were particles embedded in a rigid structure. The method was tested both using simulation and experimentation with a set of 3 robots. This approach is capable of achieving high precision movement which is fault tolerant and exhibits graceful degradation of performance. The method is highly flexible in the kinds of geometric formations that can be maintained. The principles of formation control are also useful in many other applications, including realistic computer animation of multiagent systems such as schools of fish or flocks of birds.

Movies: To the tune of the Blue Danube Quicktime Cinepak 1.3MB Quicktime Sorensen 751KB 

K.-H. Tan and M. A. Lewis. Virtual Structures for Cooperative Mobile Robot Control. In proceedings IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 1996. IEEE Xplore Citation PDF Cited by

M. A. Lewis and K.-H. Tan. High Precision Formation Control of Mobile Robots Using Virtual Structures. With M. A. Lewis. Autonomous Robots, 4(4):387-403. 1997. Kluwer Online Citation Cited by

J. L. Zhen, M. A. Lewis, and K.-H. Tan. Towards Universal Access to Robotic Resources. In proceedings IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 1996. IEEE Xplore Citation

Multiview Mirror Pyramid Panoramic Cameras

A mirror pyramid consists of a set of flat mirror faces arranged around an axis of symmetry, inclined to form a pyramid. By strategically positioning a number of conventional cameras around a mirror pyramid, the viewpoints for the individual cameras’ mirror images can be colocated at a single point within the pyramid, effectively forming a virtual camera with a wide field of view. Mirror pyramid-based panoramic cameras have a number of attractive properties, including single-viewpoint imaging, high resolution, and video rate capture. We propose a method for generalizing existing designs such that multiple viewpoints can be created in a single mirror pyramid. This enables simultaneous multiview panoramic video rate imaging with a compact design. more

Powerpoint zip

K.-H. Tan, H. Hua, and N. Ahuja. Multiview Mirror Pyramid Cameras. IEEE Transactions in Pattern Analysis and Machine Intelligence. July 2004 (Vol. 26, No. 7) pp. 941-946. PDF IEEE

K.-H. Tan, H. Hua and N. Ahuja. Multiview mirror pyramid-based panoramic cameras. In proceedings IEEE Workshop on Omnidirectional Vision (Omnivis), held in conjunction with ECCV 2002. PDF

Appearance-based Eye Tracking

The ability to detect the presence of visual attention from human users, and/or determine what a human user is looking at by estimating the direction of eye gaze is useful in many applications. We propose a new method for estimating eye gaze direction based on appearance manifolds. Using an enhancement to the appearance manifold method by proposing a nearest manifold point search technique that exploits the topological information inherently present in the manifold model, we have found that the algorithm is capable of estimating eye gaze with a mean angular error of 0.38 degrees, which is comparable to that obtained by commercially available eye trackers.

K.-H. Tan, D. Kriegman and N. Ahuja. Appearance-based Eye Gaze Estimation. In proceedings IEEE Workshop on Applications of Computer Vision (WACV) 2002. PDF

Visualizing Point Clouds with S-Shapes

Data visualization from a point set by estimating the underlying region is a problem of considerable practical interest and is an associated problem of set estimation. The most important issue in set estimation is consistency. Only a few existing point pattern shape descriptors that estimate the underlying region are consistent set estimators (a set estimator is consistent if it converges—in an appropriate sense—to the original set as the sample size increases). On the other hand, to be used as a shape descriptor, a set estimator should also satisfy several important criteria such as correct identification of number of components, robustness in the presence of noise and computational efficiency. Here we propose such a class of set estimators called s-shapes, which remain consistent in finite dimensions when the data are generated from any continuous distribution. These set estimators can be easily computed and effectively used for fast data visualization.

A. Ray Chaudhuri, A. Basu, K.-H. Tan, S. Bhandari, and B. B. Chaudhuri. An Efficient Approach to Consistent Set Estimation in Finite Dimensions. Computer Vision and Image Understanding. Accepted 2003. PDF