Snakes: Active contour models

@article{Kass2004SnakesAC,
  title={Snakes: Active contour models},
  author={Michael Kass and Andrew P. Witkin and Demetri Terzopoulos},
  journal={International Journal of Computer Vision},
  year={2004},
  volume={1},
  pages={321-331}
}
A snake is an energy-minimizing spline guided by external constraint forces and influenced by image forces that pull it toward features such as lines and edges. Snakes are active contour models: they lock onto nearby edges, localizing them accurately. Scale-space continuation can be used to enlarge the capture region surrounding a feature. Snakes provide a unified account of a number of visual problems, including detection of edges, lines, and subjective contours; motion tracking; and stereo… 

Figures from this paper

Locating object contours in complex background using improved snakes
A semi-automatic system for edge tracking with snakes
TLDR
An algorithm is presented that uses a block matching technique to guide the endpoints of the snake, optical flow to push the snake in the direction of the underlying motion, followed by the traditional snake edge-fitting minimization process for tracking facial features of an actor for driving computer animated characters.
Efficient Contour Detection Based On Improved Snake Model
  • F. Shih, Kai Zhang
  • Computer Science, Physics
    Int. J. Pattern Recognit. Artif. Intell.
  • 2004
TLDR
This paper presents a new external force field, named gravitationforce field, for the snake model, and associates this force field with edge preserving smoothing to drive the snake for solving the problems.
A NOVELL ALGORITHM FOR ACTIVES CONTOUR MODELS: ADAPTIVE SNAKES
This work presents a novel algorithm for optimizing deformable contour models. It overcomes an inherent problem of GVF snakes using a dynamic intervention at the curve evolution in a way that is
An Initialization of Active Contour Models(Snakes) using Convex Hull Approximation
TLDR
This paper proposes a new initialization of Snakes and GVF using convex hull approximation, which initializes the vertex of Snake andGVF as a convex polygonal contour near object edges.
Directional gradient vector flow for snakes
  • J. Cheng, S. Foo, S. Krishnan
  • Computer Science
    Proceedings of the Fourth IEEE International Symposium on Signal Processing and Information Technology, 2004.
  • 2004
TLDR
Experimental results demonstrate that the DGVF snake provides a better segmentation than GVF snake in situations when edges of different directions are present and may pose confusion for segmentation.
Dynamic directional gradient vector flow for snakes
TLDR
Experimental results demonstrate that the DDGVF snake provides a much better segmentation than GVF snake in situations when edges of different directions are present which pose confusion for segmentation.
Guided energy-minimizing model for segmentation of vector fields
Active contours or snakes, are a group of image segmentation methods based on the idea of energy-minimizng curves. In this paper classical snake model with added Balloon Force is modified, granting
Dividing snake algorithm for multiple object segmentation
TLDR
This work further develops the segmented snake approach to contour detection and illustrates its flexibility by showing how it can be adapted to yield a dividing snake algorithm for use in multiple object segmentation.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 49 REFERENCES
Filling-in the gaps: The shape of subjective contours and a model for their generation
TLDR
It is shown that shapes meeting all the above requirement can be generated by a network which performs simple, local computations, and it is suggested that the filling-in process plays an important role in the early processing of visual information.
Signal matching through scale space
TLDR
The general signal-matching problem is formulated as the minimization of an energy measure that combines a smoothness term and a similarity term that reduces to a dynamic system governed by a set of coupled, first-order differential equations.
Symmetry-seeking models and 3D object reconstruction
TLDR
The forces proposed in this paper deform the model in space so that the shape of its projection into the image plane is consistent with the 2D silhouette of an object of interest.
An application of heuristic search methods to edge and contour detection
TLDR
Experimental results show how various information about the shape of the contour of an object can be embedded in the figure of merit, thus allowing the extraction of contours from noisy pictures and the separation of touching objects.
Regularization of Inverse Visual Problems Involving Discontinuities
  • Demetri Terzopoulos
  • Mathematics
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 1986
TLDR
This paper proposes a general class of controlled-continuity stabilizers which provide the necessary control over smoothness in visual reconstruction problems that involve both continuous regions and discontinuities, for which global smoothness constraints fail.
Theory of edge detection
  • D. Marr, E. Hildreth
  • Mathematics
    Proceedings of the Royal Society of London. Series B. Biological Sciences
  • 1980
TLDR
The theory of edge detection explains several basic psychophysical findings, and the operation of forming oriented zero-crossing segments from the output of centre-surround ∇2G filters acting on the image forms the basis for a physiological model of simple cells.
Elastic Matching of Line Drawings
  • D. Burr
  • Engineering
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 1981
TLDR
Examples are shown on handprint and animation, suggesting the use of dynamic matching for shape recognition and for motion correspondence, and extending the model to line drawings.
Computational vision and regularization theory
TLDR
Descriptions of physical properties of visible surfaces, such as their distance and the presence of edges, must be recovered from the primary image data and algorithms and parallel analog circuits that can solve ‘ill-posed problems’ and which are suggestive of neural equivalents in the brain are proposed.
An Application of Relaxation Labeling to Line and Curve Enhancement
TLDR
A relaxation process is described and is applied to the detection of smooth lines and curves in noisy, real world images, effective even for curves of low contrast, and even when many curves lie close to one another.
The computation of the velocity field
  • E. Hildreth
  • Physics
    Proceedings of the Royal Society of London. Series B. Biological Sciences
  • 1984
TLDR
Empirical studies show the predictions of this computation of an instantaneous two-dimensional velocity field from the changing image to be consistent with human motion perception, and an additional constraint of smoothness of the velocity field is formulated, based on the physical assumption that surfaces are generally smooth.
...
1
2
3
4
5
...