原书出处
Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)
However, there are several reasons why we might not be able to apply this definition: We only have access to a sampled version of the input image, ℓ [n, m], and we can not compute the limit when ϵ goes to zero. The image could contains many nonderivable points and the gradient would not be defined. We will see how to address this issue later
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/c018.xhtml
EPUB SHA-256:60a7eebc43f5d9a675d776fed3736fbdaa791e55877023f2f8cba016a9b384f8