AI Paper Atlas

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

Unbiased Manifold Augmentation for Coarse Class Subdivision

Baoming Yan, Ke Gao, Bo Gao, Lin Wang, Jiang Yang, Xiaobo Li

ECCV 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 1 citations (OpenAlex)

Read the paperCheck 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 UMA (same-class layer-switch synthesis, UMA A+B) over best prior augmentationbest baseline augmentation · 11 CCS tasks (32 samples), averageaccuracy (%): 7.93% average improvementImproved
add UMA++ (cross-class synthesis plus robust learning) over best prior augmentationbest baseline augmentation · 11 CCS tasks (32 samples), averageaccuracy (%): 10.37% average improvementImproved
add UMA C/D cross-class synthesis with progressive robust learning (UMA++ vs UMA) (ablation)UMA · 11 CCS tasks (32 samples), averageaccuracy (%): 79.37 -> 81.81Improved
add UMA C/D cross-class synthesis with progressive robust learning (UMA++ vs UMA) (ablation)UMA · CelebA Eyeglasses (32 samples)accuracy (%): 90.99 -> 93.01Improved
add UMA++ over best prior augmentation on eyeglasses recognitionbest baseline augmentation · CelebA Eyeglasses (32 samples)accuracy (%): 14.15% over best method (12.13% for UMA)Improved
add UMA++ to ResNet-18 baseline under severe biasResNet-18 baseline · CelebA gender with eyeglasses biasaccuracy (%): 66.02 -> 82.81Improved
UMA++ vs state-of-the-art augmentation under severe biasstate-of-the-art augmentation · CelebA gender with eyeglasses biasaccuracy (%): 9.37% higherImproved
add UMA++ to SPS FGVC methodSPS · CelebA gender with eyeglasses biasaccuracy (%): 72.27 -> 83.59Improved
add UMA++ to ProtoTree FGVC methodProtoTree · CelebA gender with eyeglasses biasaccuracy (%): 66.80 -> 84.38Improved
add UMA++ to CAL FGVC methodCAL · CelebA gender with eyeglasses biasaccuracy (%): 73.05 -> 84.77Improved
use full training data instead of 32 samples, UMA vs othersUMA · sub-class recognition tasks (full training data)accuracy (%): 2% improvementImproved
use 12.5% of training data, UMA vs othersUMA · smiling and heavy makeup+smiling recognitionaccuracy (%): up to 7% (smiling), 9.8% (CA#1)Improved

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