原书出处
Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)
K-means tries to find clusters in this scatter plot (and, if you prefer a probabilistic interpretation, it can be thought of as fitting a mixture of Gaussians to the datapoints, which aims to approximate the distribution from which these pixels were sampled). We can segment via k-means by simply assigning each pixel to the cluster assignment of its color value. Below we color each pixel
原书定位详情
来源:a19c7933-4eda-4176-aebd-30fe7e1ad292
正文位置:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c031.xhtml
EPUB SHA-256:60a7eebc43f5d9a675d776fed3736fbdaa791e55877023f2f8cba016a9b384f8