Rosenfeld, A.,
Hummel, R.A., and
Zucker, S.W.,
Scene Labeling by Relaxation Operations,
SMC(6), No. 6, June 1976, pp. 420-433.
Relaxation, Early work.
Early discussion of relaxation operations. All the others derive
from this one in some way.
The basic simple updating formula.
BibRef
7606
Rosenfeld, A.,
Iterative Methods in Image Analysis,
PR(10), No. 3, 1978, pp. 181-187.
Elsevier DOI
Survey, Relaxation.
Relaxation, Survey.
BibRef
7800
Earlier:
PRIP77(14-18).
BibRef
And:
Relaxation Methods in Image Processing and Analysis,
ICPR78(181-185).
BibRef
Hummel, R.A., and
Zucker, S.W.,
On the Foundations of Relaxation Labeling Processes,
PAMI(5), No. 3, May 1983, pp. 267-287.
BibRef
8305
RCV87(585-605).
BibRef
Earlier:
ICPR80(50-53),
BibRef
McGill Univ. TRTR-80-7 is an expanded version.
Relaxation, Theory.
This paper is an attempt to develop an underlying theory for the
continuous relaxation process. One result is the development of an
explicit function to maximize to guide the relaxation process,
leading to a new relaxation operator. The second result is that
finding a consistent labeling is equivalent to solving a
variational inequality. The new operator accomplishes both goals,
and can be approximated by the more standard ones. This leads to a
conjecture that the successful application of standard operators
can be explained by this theory. This resulting technique is
similar in intent to [
See also Improving Consistency and Reducing Ambiguity in Stochastic Labeling: An Optimization Approach. ], but differs in the resulting
formulation.
BibRef
Mohammed, J.L.[John L.],
Hummel, R.A.[Robert A.], and
Zucker, S.W.[Steven W.],
A Gradient Projection Algorithm for Relaxation Methods,
PAMI(5), No. 3, May 1983, pp. 330-332.
Companion paper of
See also On the Foundations of Relaxation Labeling Processes. How the projection is computed.
BibRef
8305
Zucker, S.W.,
Leclerc, Y.G., and
Mohammed, J.L.,
Continuous Relaxation and Local Maxima Selection:
Conditions for Equivalence,
PAMI(3), No. 2, March 1981, pp. 117-127.
BibRef
8103
Earlier:
IJCAI79(1014-1016).
BibRef
Haralick, R.M.,
Mohammed, J.L.[John L.],
Zucker, S.W.,
Compatibilities and the Fixed Points of Arithmetic Relaxation Processes,
CGIP(13), No. 3, July 1980, pp. 242-256.
Elsevier DOI
BibRef
8007
Parent, P., and
Zucker, S.W.,
Radial Projection: An Efficient Update Rule for Relaxation Labeling,
PAMI(11), No. 8, August 1989, pp. 886-889.
IEEE DOI The projection is radial, not normal, to avoid
problems at the boundaries of the label space.
BibRef
8908
Zucker, S.W.,
Relaxation Labelling and the Reduction of Local Ambiguities,
ICPR76(852-861).
BibRef
7600
And:
CMetImAly77(445-454).
BibRef
And:
Relaxation Labelling, Local Ambiguity, and Low-Level Vision,
PRAI-76(593-616)
BibRef
Zucker, S.W.,
Toward Consistent Descriptions in Vision Systems,
IJCAI77(709).
BibRef
7700
Zucker, S.W., and
Mohammed, J.L.,
Analysis of Probabilistic Relaxation Labeling Processes,
PRIP78(307-312).
BibRef
7800
Zucker, S.W.,
Leclerc, Y.G., and
Mohammed, J.L.,
Continuous Relaxation Local Maxima Selection, Conditions for
Equivalence,
IJCAI79(1014-1016).
BibRef
7900
And:
McGill Univ. TRTR-78-157, December 1978.
Theoretical analysis of relaxation (a lot of clarity is lost in the
short paper) if all maximal nodes support one another then this is where the
relaxation ends up.
BibRef
Hummel, R.A.,
Rosenfeld, A.,
Relaxation Processes for Scene Labeling,
SMC(8), 1978, pp. 765-768.
BibRef
7800
Zucker, S.W.,
Krishnamurthy, E., and
Haar, R.L.,
Relaxation Processes for Scene Labeling:
Convergence, Speed, and Stability,
SMC(8), No. 1, January 1978, pp. 41-48.
BibRef
7801
Haar, R.L.[Robert L],
Sketching: Estimating Object Positions from Relational Descriptions,
CGIP(19), No. 3, July 1982, pp. 227-247.
Elsevier DOI
BibRef
8207
Zucker, S.W.,
Relaxation Labeling: 25 Years and Still Iterating,
FIU01(Chapter 10).
BibRef
0100
Chapter on Matching and Recognition Using Volumes, High Level Vision Techniques, Invariants continues in
Shmuel Peleg Theoretical Relaxation Papers .