AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

MetaSAug: Meta Semantic Augmentation for Long-Tailed Visual Recognition

Shuang Li, Kaixiong Gong, Chi Harold Liu, Yulin Wang, Feng Qiao, Xinjing Cheng

CVPR 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Decoupled and calibrated long-tail training · 174 citations (OpenAlex)

Check your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add re-weighting to MetaSAug (ablation)MetaSAug with cross-entropy loss · CIFAR-LT-10, imbalance factors 100, 50, 20top-1 error (%): 25.96 -> 19.46; 20.63 -> 15.97; 15.15 -> 12.36Improved
Add meta-learning to learn class-wise covariance (ablation)MetaSAug with cross-entropy loss · CIFAR-LT-10, imbalance factors 100, 50, 20top-1 error (%): 21.68 -> 19.46; 17.43 -> 15.97; 13.08 -> 12.36Improved
Learn covariance with MetaSAug rather than using fixed ISDA with L2RW (ablation)CIFAR-LT-10 cross-entropy training · CIFAR-LT-10, imbalance factors 100, 50, 20top-1 error (%): 25.16 -> 19.46; 20.78 -> 15.97; 16.53 -> 12.36Improved
Learn covariance with MetaSAug rather than using fixed ISDA with Meta-weight-net (ablation)CIFAR-LT-10 cross-entropy training · CIFAR-LT-10, imbalance factors 100, 50, 20top-1 error (%): 24.69 -> 19.46; 20.42 -> 15.97; 14.71 -> 12.36Improved
Learn covariance with MetaSAug rather than using fixed ISDA with meta-class-weight (ablation)CIFAR-LT-10 cross-entropy training · CIFAR-LT-10, imbalance factors 100, 50, 20top-1 error (%): 20.78 -> 19.46; 17.12 -> 15.97; 12.93 -> 12.36Improved

Later papers that compared with this method

2 later accepted papers report MetaSAug as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy