AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

RDA: Robust Domain Adaptation via Fourier Adversarial Attacking

Jiaxing Huang, Dayan Guan, Aoran Xiao, Shijian Lu

ICCV 2021 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Adversarial domain adaptation · 75 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 FAA on supervised source loss only (FAA-S) (ablation)self-training (ST) · GTA5 to CityscapesmIoU: 42.6 -> 47.8Improved
add FAA on unsupervised target loss only (FAA-T) (ablation)self-training (ST) · GTA5 to CityscapesmIoU: 42.6 -> 48.5Improved
add FAA on both source and target losses vs FAA-T only (ablation)self-training (ST) · GTA5 to CityscapesmIoU: 48.5 -> 50.1Improved
add FAA on both losses vs FAA-T only (ablation)AdaptSeg · GTA5 to CityscapesmIoU: 47.1 -> 48.0Improved
add FAA to supervised-only baselinesource-only baseline · GTA5 to CityscapesmIoU: 36.6 -> 45.2Improved
add FAACRST · SYNTHIA to CityscapesmIoU*: 50.1 -> 55.9Improved
add FAACRST · VisDA17mean accuracy: 78.1 -> 82.7Improved
add FAATAT · Office-31mean accuracy: 88.4 -> 90.3Improved
add FAASWDA · Cityscapes to Foggy CityscapesmAP: 34.3 -> 38.3Improved
add FAACRDA · Cityscapes to BDD100kmAP: 26.9 -> 29.9Improved
add dropout to self-training (ablation)ST · GTA5 to CityscapesmIoU: 42.6 -> 43.1Improved
add mixup to self-training (ablation)ST · GTA5 to CityscapesmIoU: 42.6 -> 43.6Improved

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