Constraints on deformable models: recovering 3D shape and nongrid motion
Artificial Intelligence
Active vision
Active vision
SIAM Journal on Numerical Analysis
International Journal of Computer Vision
Minimal Surfaces Based Object Segmentation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Deformable template models: a review
Signal Processing - Special issue on deformable models and techniques for image and signal processing
The velocity snake: deformable contour for tracking in Spatio-Velocity space
Computer Vision and Image Understanding
Geodesic Active Contours and Level Sets for the Detection and Tracking of Moving Objects
IEEE Transactions on Pattern Analysis and Machine Intelligence
Robot Vision
IEEE Transactions on Pattern Analysis and Machine Intelligence
Front Propagation and Level-Set Approach for Geodesic Active Stereovision
ACCV '98 Proceedings of the Third Asian Conference on Computer Vision-Volume I - Volume I
Stereo Coupled Active Contours
CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
NAM '97 Proceedings of the 1997 IEEE Workshop on Motion of Non-Rigid and Articulated Objects (NAM '97)
Geometric Partial Differential Equations and Image Analysis
Geometric Partial Differential Equations and Image Analysis
Variational principles, surface evolution, PDEs, level set methods, and the stereo problem
IEEE Transactions on Image Processing
IEEE Transactions on Image Processing
Hi-index | 0.00 |
This paper presents a 3D active contour model for boundary detection and tracking of non-rigid objects, which applies stereo vision and motion analysis to the class of energy-minimizing deformable contour models, known as snakes. The proposed contour evolves in three-dimensional space in reaction to a 3D potential function, which is derived by projecting the contour onto the 2D stereo images. The potential function is augmented by a kinetic term, which is related to the velocity field along the contour. This term is used to guide the inter-image contour displacement. The incorporation of inter-frame velocity estimates in the tracking algorithm is especially important for contours which evolve in 3D space, where the added freedom of motion can easily result in loss of tracking. The proposed scheme incorporates local velocity information seamlessly in the snake model, with little computational overhead, and does not require exogenous computation of the optical flow or related quantities in each image. The resulting algorithm is shown to provide good tracking performance with only one iteration per frame, which provides a considerable advantage for real time operation.