Boundary finding with parametrically deformable models
read more
Citations
Active appearance models
Active Appearance Models
Deformable models in medical image analysis: a survey
Robot vision
Markov Random Field Modeling in Image Analysis
References
A mathematical theory of communication
Snakes : Active Contour Models
Use of the Hough transformation to detect lines and curves in pictures
Related Papers (5)
Frequently Asked Questions (19)
Q2. How is the surface expected to be distinguishable from the background?
The surface is expected to be distinguishable by some measure of boundary strength (direction can also be used) computed from the image.
Q3. What are the main advantages of a superquadric?
Hyperquadrics [24] are a generalization of superquadrics that allow smooth deformations from shapes with convex polyhedral bounds, although no explicit parametrized form is possible.
Q4. What is the reason why the global optimum is not necessary?
Since a local optimization method is likely to be sufficient, although there is still the possibility of converging to a poor local maximum, the excessive computation involved in finding a global optimum is deemed not necessary.
Q5. How can curvature be calculated without an analytic description of the surface?
Without an analytic description of the surface, curvature can be calculated based on the computation of derivatives from a local surface patch fit, or from a discrete approximation of the derivatives at each point.
Q6. What are the main advantages of a generalized cylinder?
If the spine and crosssection are represented parametrically, as opposed to directly as an explicit list of coordinates or segments, generalized cylinders can be completely parametric.
Q7. What is the disadvantage of the Hough method?
the storage and computational complexity of the Hough method are a great disadvantage, especially if deformations are envisaged.
Q8. What is the definition of a superquadric?
Superquadrics are an extension of quadrics using an exponent that allows the shape to vary from an ellipsoid to a rectangular parallelepiped.
Q9. How is the model matched to the image?
The model is matched to the image by optimizing in the parameter space the match between the model and a boundary measure applied to the image.
Q10. What is the objective function for deformable object boundary finding?
The deformable object boundary finding method has been applied to a variety of objects from real images, with an emphasis on heart and brain images using primarily magnetic resonance images.
Q11. What is the recent work done on deformable templates?
Work has also been done developing deformable templates based on Markov models of two-dimensional boundaries incorporating knowledge of shape from statistical features [16].
Q12. What is the measure of fit for curves?
The measure of fit for curves can be written as follows, here using only boundary strength:M(b,p) =∫ S0|b(x(p, s), y(p, s), z(p, s))|ds (22)where p is the vector consisting of the basis function parameters.
Q13. What are the properties of the ellipse?
Each ellipse can be described by four geometric properties: semi-major axis length, semi-minor axis length, rotation and phase shift.
Q14. Why is the surface description so straightforward?
The Fourier surface description makes the calculation of geometric surface properties straightforward because a continuous description of the surface is known.
Q15. Why is this only done for the curve models?
This has been done only for the curve models so far because invariance to the surface parametrization has not been established for the surface models.
Q16. What is the way to determine the K?
Although this implies fixing the highest order harmonic used, an iterative method for determining the best K using a trade-off between conciseness and fit could be devised.
Q17. What is the equivalent measure for surfaces?
The equivalent measure for surfaces is:M(b,p) =∫ ∫A|b(x(p, u, v), y(p, u, v), z(p, u, v))|dA (23)Equation 23 can be evaluated by numerical integration.
Q18. What is the definition of a dynamic, three-dimensional imaging device?
The DSR is a dynamic, three-dimensional imaging device based on high-speed x-ray computed tomographycapable of imaging the moving heart [34].
Q19. What are the low-order frequency displacement eigenvectors for a shape?
Shapes are represented by the low-order frequency displacement eigenvectors corresponding to the free vibration modes of the object.