AI Paper Atlas

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

Domain transfer through deep activation matching

Haoshuo Huang, Qixing Huang, Philipp Krahenbuhl

ECCV 2018 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Adaptation for semantic segmentation · 111 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 weight-drift regularization to GAM (ablation)GAM · SVHN->MNISTaccuracy: 73.6 -> 74.6Improved
add weight-drift regularization to GAM (ablation)GAM · MNIST->USPSaccuracy: 95.6 -> 95.7No clear effect
add our regularization to ADDAADDA · SVHN->MNISTaccuracy: 72.0 -> 73.5Improved
add our regularization to ADDAADDA · GTA->CityscapesIoU: 28.6 -> 30.5Improved
add Gaussian i.i.d. activation matchingERFNet · GTA->CityscapesIoU: 15.8 -> 28.0Improved
replace Gaussian matching with GAN-based general activation matching (ablation)GAM · GTA->CityscapesIoU: 28.0 -> 31.3Improved
replace ADDA with GAMERFNet · GTA->CityscapesIoU: 28.6 -> 31.3Improved
replace CyCADA with GAMDilated Residual Network · GTA->CityscapesIoU: 39.5 -> 40.2Improved
adapt SYNTHIA-trained model with GAMERFNet · SYNTHIA->CityscapesIoU: 45.19 -> 62.26Improved
finetune GAM from Curriculum modelCurriculum · SYNTHIA->CityscapesIoU: 29.0 -> 30.7Improved
skew target label distribution (full SVHN source, odd MNIST target)GAM · SVHN->MNISTaccuracy: 50.8 -> 29.8Hurt
adapt with GAM when source skewed (odd SVHN to full MNIST)GAM · SVHN->MNISTaccuracy: 32.5 -> 36.9Improved

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