Using Dynamic Programming for Solving Variational Problems in Vision
IEEE Transactions on Pattern Analysis and Machine Intelligence
Active vision
Catching moving objects with snakes for motion tracking
Pattern Recognition Letters
Generalized gradient vector flow external forces for active contours
Signal Processing - Special issue on deformable models and techniques for image and signal processing
Geodesic Active Contours and Level Sets for the Detection and Tracking of Moving Objects
IEEE Transactions on Pattern Analysis and Machine Intelligence
Active Contours: The Application of Techniques from Graphics,Vision,Control Theory and Statistics to Visual Tracking of Shapes in Motion
Tracking Deformable Objects in the Plane Using an Active Contour Model
IEEE Transactions on Pattern Analysis and Machine Intelligence
Region Tracking via Level Set PDEs without Motion Computation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Region tracking through image sequences
ICCV '95 Proceedings of the Fifth International Conference on Computer Vision
Level set analysis for leukocyte detection and tracking
IEEE Transactions on Image Processing
Constraining active contour evolution via Lie Groups of transformation
IEEE Transactions on Image Processing
Hi-index | 0.01 |
We investigate the evolution of active contours in terms of progressive modification of an initial contour following the chosen Lie group of object-to-image transformations. Because of non-fronto-parallel viewing of an object or due to relative motion between the camera and the object, the resultant image may undergo affine or projective object-to-image transformations. In a recent paper we have shown that in the case of object tracking, frame-to-frame deformations of an initial curve obtained through Euler-Lagrange descent equations of a curve functional can be used to enact a desired Lie group of plane transformations [A.-R. Mansouri, D.P. Mukherjee, S.T. Acton, Constraining active contour evolution via Lie groups of transformation, IEEE Trans. Image Process. 13 (2004) 853-863]. In this work, we propose an energy functional that encodes the Lie group transformation parameters, which in turn guide shape distortion due to oblique viewing. Additional constraints, such as transformation smoothness, are imposed on the active contour by modifying the energy functional. The functional is minimized using numerical schemes similar to the conjugate gradient technique, and the convergence properties are discussed. The success of the technique for affine and projective scenes is demonstrated with both synthetic and real image examples and compared with the related approaches.