Uncertainty optimization for robust dynamic optical flow estimation

We develop an optical flow estimation framework that focuses on motion estimation over time formulated in a dynamic Bayesian network. It realizes a spatiotemporal integration of motion information using a dynamic and robust prior that incorporates spatial and temporal coherence constraints on the flow field. The main contribution is the embedding of these particular assumptions on optical flow evolution into the Bayesian propagation approach that leads to a computationally feasible two-filter inference method and is applicable for on and offline parameter optimization. We analyse the possibility to optimize imposed Student's t-distributed model uncertainties, which are the camera noise and the transition noise. Experiments with synthetic sequences illustrate how the probabilistic framework improves the optical flow estimation because it allows for noisy data, motion ambiguities and motion discontinuities.

[1]  Alfred Mertins,et al.  Flexible, highly scalable, object-based wavelet image compression algorithm for network applications , 2004 .

[2]  Alexandru Nicolau,et al.  Computing Programs Containing Band Linear Recurrences on Vector Supercomputers , 1996, IEEE Trans. Parallel Distributed Syst..

[3]  Stuart J. Russell,et al.  Dynamic bayesian networks: representation, inference and learning , 2002 .

[4]  G. Kitagawa.  The two-filter formula for smoothing and an implementation of the Gaussian-sum smoother , 1994 .

[5]  Susanne E. Hambrusch,et al.  Parallel scalable libraries and algorithms for computer vision , 1994, Proceedings of the 12th IAPR International Conference on Pattern Recognition, Vol. 2 - Conference B: Computer Vision & Image Processing. (Cat. No.94CH3440-5).

[6]  Gert Cauwenberghs,et al.  Hybrid support vector machine/hidden Markov model approach for continuous speech recognition , 2000, Proceedings of the 43rd IEEE Midwest Symposium on Circuits and Systems (Cat.No.CH37144).

[7]  Michael J. Black,et al.  On the Spatial Statistics of Optical Flow , 2005, ICCV.

[8]  Rudolf Mester,et al.  A Maximum Likelihood Estimator for Choosing the Regularization Parameters in Global Optical Flow Methods , 2006, 2006 International Conference on Image Processing.

[9]  Alan L. Yuille,et al.  Probabilistic Motion Estimation Based on Temporal Coherence , 2000, Neural Computation.

[10]  David J. Fleet,et al.  Performance of optical flow techniques , 1994, International Journal of Computer Vision.

[11]  Jieping Ye,et al.  A two-stage linear discriminant analysis via QR-decomposition , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  J. Weickert,et al.  Lucas/Kanade meets Horn/Schunck: combining local and global optic flow methods , 2005 .

[13]  Young Cho,et al.  Scalable network based FPGA accelerators for an automatic target recognition application , 1998, Proceedings. IEEE Symposium on FPGAs for Custom Computing Machines (Cat. No.98TB100251).

[14]  Edward H. Adelson,et al.  Probability distributions of optical flow , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[15]  Maria Gabrani,et al.  Scalable algorithms for media processing , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[16]  A. Jepson,et al.  Sparse PCA. Extracting multi-scale structure from data , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[17]  William M. Campbell,et al.  Speaker recognition with polynomial classifiers , 2002, IEEE Trans. Speech Audio Process..

[18]  Michael J. Black,et al.  Robust dynamic motion estimation over time , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[19]  Patrick Bouthemy,et al.  Multimodal Estimation of Discontinuous Optical Flow using Markov Random Fields , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[20]  Anuj Srivastava,et al.  Optimal linear representations of images for object recognition , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[21]  Qiang Yang,et al.  A Novel Scalable Algorithm for Supervised Subspace Learning , 2006, Sixth International Conference on Data Mining (ICDM'06).

[22]  Amitabha Das,et al.  Estimation of Occlusion and Dense Motion Fields in a Bidirectional Bayesian Framework , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[23]  Volker Willert,et al.  Non-Gaussian velocity distributions integrated over space, time, and scales , 2005, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[24]  Eero P. Simoncelli,et al.  Noise characteristics and prior expectations in human visual speed perception , 2006, Nature Neuroscience.

[25]  Ajit Singh,et al.  Incremental estimation of image-flow using a Kalman filter , 1991, Proceedings of the IEEE Workshop on Visual Motion.