en-de
en-es
en-fr
en-sl
en
en-zh
0.25
0.5
0.75
1.25
1.5
1.75
2
Learning issues in image segmentation
Published on Feb 25, 200714686 Views
Image segmentation is often defined as a partitioning of pixels or image blocks into homogeneous groups. These groups are characterized by a prototypical vector in feature space, e.g., the space of Ga
Related categories
Chapter list
Learning Issues in & Image Segmentation00:00
The Problem of Data Clustering00:50
Application: Image segmentation03:09
Clustering Approach to Image Segmentation06:42
Problem Formalization07:24
Cost Function Idea08:44
Cost Function-Based Clustering12:00
Cost Function Optimization15:18
Extension16:48
Unsupervised vs. supervised18:51
What is Data Clustering?20:17
Generalization Problem in Classification21:19
Structure of the Tutorial23:46
Data Types in Clustering Problems25:15
Part I: Clustering Principles26:44
Vectorial Data:30:27
k-Means Problem30:39
k-Means Algorithm32:12
k-Means Segmentation of LANDSAT Images32:46
Example Mixture Model35:38
Parametric Distributional Clustering36:04
Parametric Distributional Clustering36:31
Parametric Distributional Clustering36:34
Parametric Distributional Clustering37:05
Gaussian Mixture Models37:37
Gaussian Mixture Models38:25
Gaussian Mixture Models38:53
Gaussian Mixture Models39:01
Gaussian Mixture Models39:03
Gaussian Mixture Models39:04
Gaussian Mixture Models39:09
Gaussian Mixture Models39:12
Gaussian Mixture Models39:14
Gaussian Mixture Models39:19
Gaussian Mixture Models39:20
Gaussian Mixture Models39:22
Gaussian Mixture Models39:30
Generative Model39:53
Maximum Likelihood Approach40:35
Cost Function for PDC41:43
Information Bottleneck43:30
Information Bottleneck45:04
PDC Segmentation46:15
PDC Segmentation46:20
PDC Resampling47:06
SAR Imagery47:39
Proximity Data47:59
Proximity Data: Example55:04
Proximity Data in Segmentation55:11
The Pairwise Clustering Cost Function56:46
Invariance Properties of Hpc58:21
Constant Shift Embedding59:57
Clustering of Bacterial GyrB Sequences01:03:00
Globin Proteins: Cluster Solution01:03:49
Normalized Cut01:03:55
Relaxation of NCut01:06:02
Example Normalized Cut01:07:26
Part II: Optimization Methods01:08:47
The Maximum Entropy Principle01:10:19
Metropolis Sampler for Clustering01:11:15
Algorithm Design for Maximum Entropy Estimation01:12:42
Phase Transitions in K-means Clustering01:13:23
Cooling Dynamics of PDC01:14:36
Cooling Dynamics of PDC01:15:02
Cooling Dynamics of PDC01:15:29
Cooling Dynamics of PDC01:15:30
Cooling Dynamics of PDC01:15:32
Cooling Dynamics of PDC01:15:41
Cooling Dynamics of PDC01:15:42
Cooling Dynamics of PDC01:15:42
Phase Transitions in Segmentation01:15:59
Part III: Cluster Validation01:16:50
The Problem of Cluster Validity01:18:50
Validation Methods ...01:19:52
Complexity-based Validation01:21:00
Underlying Principle01:21:31
BIC Validation of a Mixture01:22:18
Stability-based Validation01:22:44
Stability01:24:24
Two Sample Scenario01:25:15
Measuring disagreement01:26:36
Stability Measure: Labelings on disjoint sets01:27:48
Stability Measure: Breaking Permutation Symmetry01:29:32
Stability Measure: Different Values of k01:29:59
The Final Stability Measure01:30:36
Results on Toy Data01:32:12
Class Discovery01:33:42
Clustering of Globins01:34:16
Stability: Summary01:34:34