Robust Motion Estimation under Varying Illumination
The optical-flow approach has emerged as a major technique for estimating scene and object motion in image sequences. However, the problem of motion estimation, in general, is made difficult by large illumination variations and by motion discontinuities. In recent papers, we and others have proposed global approaches to deal with both problems simultaneously within the regularization framework. A major drawback of such global methods is that several regularization parameters responsible for the integration of the illumination and motion components need to be determined in advance. This has reduced the applicability of global methods. To diminish this shortcoming of the global approach, we have proposed a parameter-free local approach, which solves a linear regression problem using a simple parametric model. To achieve robustness for the linear regression problem, we introduce a modified version of the least median of squares algorithm. Our results show that our local method is comparable to the best results obtained by the global approaches yet does not require any manual selection of parameters.
The following images show some of the results obtained from our global and local methods and other methods.  For technical details of our method, we refer to the publications at the end of this page.
Figure 1.  Test Images Sequence
Image Sequence Correct Flow u in gray scale image
Random Square Sequence                                 .
Marble Block Sequence                                  .
Figure 2. Optical flow results for Random Square Sequence
Figure 3. Optical flow results for Random Square Sequence
Figure 4. Optical flow results for Real Image Sequences
    •    Yeonho Kim
    This project is partially supported by NSF Grant No. 0414953
Yeon-Ho Kim, Aleix M. Martinez and Avi C. Kak, "Robust Motion Estimation under Varying Illumination", Image and Vision Computing, vol. 23, no. 4, pp. 365-375, 2005. [pdf]
Yeon-Ho Kim, Aleix M. Martinez and Avi C. Kak, "A Local Approach for Robust Optical Flow Estimation under Varying Illumination", British Machine Vision Conference 2004, London, September, 2004. [pdf]
Y.-H Kim and A. C. Kak, "Error Analysis of Robust Optical Flow Estimation by Least Median of Squares Methods for the Varying Illumination Model," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 28, no. 9, pp. 1418-1435,  2006. [pdf]