How does CutMix differ from Mixup?

medium

Answer

$Cut a rectangular patch from image B and paste it into image A.$
$y\; = \;{\lambda} \cdot y_{A}\; + \;(1 - {\lambda}) \cdot y_{B}$
$Unlike Mixup (which blends pixel values everywhere and creates 'ghost' images), CutMix keeps local pixel statistics intact — better for localization tasks and dense prediction.$
$Often combined with Mixup in modern vision recipes.$

Check yourself — multiple choice

CutMix and Mixup are identical
Paste a rectangular patch from another image; label mixed by area — preserves local pixel stats better than Mixup
CutMix removes all pixels
Only for text

CutMix: patch paste with area-based label mix → keeps local textures intact.

#augmentation#mixup

214 interview questions in this topic.