Design and Use of Linear Models for Image Motion Analysis Export

@article{citeulike:2647281, abstract = {Linear parameterized models of optical flow, particularly affine models, have become widespread in image motion analysis. The linear model coefficients are straightforward to estimate, and they provide reliable estimates of the optical flow of smooth surfaces. Here we explore the use of parameterized motion models that represent much more varied and complex motions. Our goals are threefold: to construct linear bases for complex motion phenomena; to estimate the coefficients of these linear models; and to recognize or classify image motions from the estimated coefficients. We consider two broad classes of motions: i) generic ” motion features” such as motion discontinuities and moving bars; and ii) non-rigid, object-specific, motions such as the motion of human mouths. For motion features we construct a basis of steerable flow fields that approximate the motion features. For object-specific motions we construct basis flow fields from example motions using principal component analysis. In both cases, the model coefficients can be estimated directly from spatiotemporal image derivatives with a robust, multi-resolution scheme. Finally, we show how these model coefficients can be use to detect and recognize specific motions such as occlusion boundaries and facial expressions.}, author = {Fleet, David J. and Black, Michael J. and Yacoob, Yaser and Jepson, Allan D.}, citeulike-article-id = {2647281}, citeulike-linkout-0 = {http://dx.doi.org/10.1023/A:1008156202475}, day = {1}, doi = {10.1023/A:1008156202475}, journal = {International Journal of Computer Vision}, keywords = {affine, classification, linear, model, motion}, month = {February}, number = {3}, pages = {171--193}, posted-at = {2008-04-09 22:03:39}, priority = {2}, title = {Design and Use of Linear Models for Image Motion Analysis}, url = {http://dx.doi.org/10.1023/A:1008156202475}, volume = {36}, year = {2000} }

TY - JOUR ID - citeulike:2647281 L3 - citeulike-article-id:2647281 N2 - Linear parameterized models of optical flow, particularly affine models, have become widespread in image motion analysis. The linear model coefficients are straightforward to estimate, and they provide reliable estimates of the optical flow of smooth surfaces. Here we explore the use of parameterized motion models that represent much more varied and complex motions. Our goals are threefold: to construct linear bases for complex motion phenomena; to estimate the coefficients of these linear models; and to recognize or classify image motions from the estimated coefficients. We consider two broad classes of motions: i) generic “motion features” such as motion discontinuities and moving bars; and ii) non-rigid, object-specific, motions such as the motion of human mouths. For motion features we construct a basis of steerable flow fields that approximate the motion features. For object-specific motions we construct basis flow fields from example motions using principal component analysis. In both cases, the model coefficients can be estimated directly from spatiotemporal image derivatives with a robust, multi-resolution scheme. Finally, we show how these model coefficients can be use to detect and recognize specific motions such as occlusion boundaries and facial expressions. IS - 3 JF - International Journal of Computer Vision EP - 193 TI - Design and Use of Linear Models for Image Motion Analysis VL - 36 SP - 171 KW - affine KW - classification KW - linear KW - model KW - motion AU - Fleet, David AU - Black, Michael AU - Yacoob, Yaser AU - Jepson, Allan PY - 2000/02/01/ UR - http://dx.doi.org/10.1023/A:1008156202475 ER -