W
Nov 29, 2025Why Mixup Works (Even When It Blends Everything Into Goo)
What is Mixup? Mixup, introduced in 2018, is one of the oddest augmentations ever proposed. Instead...
Nov 29, 20252 min read0 reactions0 comments
Back to articles 
Oct 24, 2025
Mar 20, 2024
Tag archive
#augmentations
W
Nov 28, 2025Why CutMix Works (Even When It Breaks the Image Apart)
What is CutMix? CutMix, introduced in 2019, takes Cutout’s idea and dials it up: Instead of...
Nov 28, 20252 min read0 reactions0 comments
W
Nov 17, 2025Why TrivialAugment Works (Even When It Shouldn’t)
What is TrivialAugment (TA)? TrivialAugment sits at the far end of the augmentation-policy...
Nov 17, 20252 min read0 reactions0 comments
T
Nov 2, 2025The middle brother in classifier development: What is RandAugment?
For those unaware, RandAugment is the successor to AutoAugment (AA). Introduced in 2020, it works to...
Nov 2, 20252 min read0 reactions0 comments
Here's what I learnt after spending 6 months studying image augmentations.
October 2025 My last post explored what the most common pitfalls during early classifier training...
Oct 24, 20254 min read0 reactions0 comments
Chromatic Aberration Transform in Albumentations 1.4.2
Albumentations 1.4.2 adds the Chromatic Aberration transform. This feature simulates the common lens...
Mar 20, 20241 min read0 reactions0 comments