Lecture 14.4: Image Segmentation [Clustering Based]
Lecture 14.4: Image Segmentation [Clustering Based]
Outline of the Lecture:
1. Clustering based
1. K-means (SLIC)
2. Mean-shift
What is Clustering?
What is a natural grouping?
What is similarity
Distance metrics
K-means
K-means: Initialize centers randomly
K-means: assign points to nearest center
K-means: readjust centers
K-means:
K-means loss function
K-means variations/ parameters
Seed choice
SLIC
Mean-Shift segmentation
Mean-Shift basics
Mean-Shift
Practical Knowledge
Mean Shift Clustering
Trajectory: x, m(x), m(m(x)), …
Mean-Shift Segmentation
Ex: Mean-Shift Segmentation
Mean-shift clustering
Demo
Mean-shift pros and cons