Research:

Heteroscedasticity
Robust Analysis
Bootstrap
Retrieval
Miscellaneous

Code

Publications

Theses

Tutorials

Accurate modeling of images is difficult, and bootstrap can provide the tool to reduce and/or assess the influence of a priori assumptions. We have employed bootstrap mostly for the quantitative assessment of performance in image understanding tasks with real data. Covariance matrices and confidence intervals are computed for the estimated parameters and individually for the corrected data points. As an example, the proposed methodology was applied to 3D rigid motion estimation. Older publications illustrate other applications, some related to bootstrap only through the kinship of the paradigm.

Publications
Please use the link "Abstract" to see the publishing history of a paper.
The links "Paper" also contain the abstract.

B. Matei, P. Meer: Bootstrapping errors-in-variables models.
Abstract   Paper (pdf)   Paper (ps.gz)

B. Matei, P. Meer: Optimal rigid motion estimation and performance evaluation with bootstrap.
BEST STUDENT PAPER AWARD     1999 Computer Vision and Pattern Recognition Conference.
Abstract   Paper (pdf)   Paper (ps.gz)

P. Meer, B. Matei, K. Cho: Input guided performance evaluation.
Abstract   Paper (pdf)   Paper (ps.gz)

B. Matei, P. Meer and D. Tyler: Performance assessment by resampling: Rigid motion estimators.
Abstract   Paper (pdf)   Paper (ps.gz)

K. Cho, P. Meer and J. Cabrera: Performance assessment through bootstrap.
Abstract    Paper (pdf)    Paper (ps.gz)

K. Cho, P. Meer: Image segmentation from consensus information.
Abstract Paper (pdf)       Paper (ps.gz)

J. Cabrera, P. Meer: Unbiased estimation of ellipses by bootstrapping.
Abstract    Paper (pdf)    Paper (ps.gz)