Tao Zhang, Ph.D. (2005)
{zhangt3 at cs.rpi.edu}
Publication
- T. Zhang, T. Boult, Realistic Stereo Error Models and Finite Optimal Stereo Baselines. , IEEE WACV 2011.
- T. Zhang, Z. Wang and J.F. Zhai, D. Doyen, Automatic 3D video format detection. , SPIE Electronic Imaging 2011.
- S. He, T. Zhang, D. Doyen, Visual Discomfort Prediction for Stereo Contents. , SPIE Electronic Imaging 2011.
- Z.Wang, J.F. Zhai and T. Zhang, Single Image Chromatic Aberration Reduction with Robust Local Displacement Estimation. IEEE International Conference on Image Processing, Hong Kong, 2010.
- T. Zhang, 3D image format identification by image difference. IEEE International Workshop on Hot Topics in 3D (in conjunction with ICME 2010), Singapore, 2010.
- Zhe Wang, Jiefu Zhai, Tao. Zhang and Joan Llach, Interactive Tone Mapping for High Dynamic Range Video. IEEE International Conference on Acoustics,Speech and Signal Processing,Dallas, Texas, 2010.
- T. Zhang, T. Boult and R.C. Johnson, Two thresholds are better than one. The 7th IEEE Internation Workshop on Visual Surveillance (in conjunction with CVPR 2007).
- T. Zhang, D. Freedman, Improving performance of distribution tracking through background mismatch, IEEE Trans. on PAMI, 27(2):282-287, 2005.
- D. Freedman, R.Radke, T. Zhang, Y. Jeong, D.M. Lovelock and G.Y. Chen, Model-based segmentation of medical Imagery by matching distributions, IEEE Trans. on Medical Imaging, 24(3):281-292, 2005.
- D. Freedman, T. Zhang, Active contours for tracking distributions. IEEE Trans. on Image Processing, 13(4):518-526, 2004.
- D.Freedman and T. Zhang, Interactive Graph Cut Based Segmentation With Shape Priors, IEEE International Conference on CVPR, San Diego, CA, 2005.
- D. Freedman,R.Radke, T. Zhang, Y. Jeong, Model-based multi-object segmentation via distribution matching, IEEE Workshop on Articulated and Nonrigid Motion (in conjunction with CVPR 2004), Washington. D.C, 2004.
- T. Zhang and D. Freedman, Tracking objects using density matching and shape priors, the Ninth IEEE Int. Conf. on Computer Vision, Nice, France, 2003.
- R.J. Radke, D. Freedman, Y. Jeong, T. Zhang, and G.T.Y Chen. Deformable model-based segmentation of 3D CT by matching distributions. the American Association of Physicists in Medicine (AAPM) Annual Meeting, 2005.
- Tao Zhang, Zhi-hong Mao, Shao-wei Xia, A method to extract numeral string overlapping borders, Chinese Information Journal, vol 12, n2, 1998
- S. Wang, F. Ma, B. Zhao, T. Zhang, G.J. Ma, S.W.Xia, A new fault tolerant Chinese bank check recognition system, Proceeding of 1997 IEEE Int. Conf. in Intelligent Processing System, Beijing, China, 1997
Selected Patent Applications
R.D. Yang, Y. Luo, T.Zhang, Method to control media with face detection and hot spot motion, 2008
R.D. Yang, Y. Luo, T.Zhang, Method for controlling and requesting information from displaying multimedia, 2008
N. Yao, I. Izzat, T. Zhang, D.Q. Zhang, Apparatus and method for filling holes in depth maps, 2009
T. Zhang, Method and apparatus for distinguishing a 3D image from a 2D image and for identifying the presence of a 3D image format by image difference determination, 2009
T. Zhang, Method and apparatus for distinguishing a 3D image from a 2D image and for identifying the presence of a 3D image format by feature correspondence determination, 2009
T. Zhang, System and method for combining 3D text with 3D content, 2009
T. Zhang,Z. Wang and J.F. Zhai, Method for detecting frame sequential 3D video sequences, 2010
S. He and T. Zhang, Method and apparatus for customizing 3-Demensional effects of stereo content, 2010
Useful code
memJpegDecoder: In Computer vision projects, it is often desirable to decode (decompress) JPEG image in memory directly. However, IJG's jpeg library's default behavious is to decompress the image into a file. The memJpegDecoder package redefines the data source so that the JPEG image can be decompressed in memory. It is written in C/C++. It provides two interfaces. If you use OpenCV, a decompressed IplImage is returned; otherwise, you can access the decompressed image from an outputBuffer directly.