AI Paper Atlas

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

GA-DAN: Geometry-Aware Domain Adaptation Network for Scene Text Detection and Recognition

Fangneng Zhan, Chuhui Xue, Shijian Lu

ICCV 2019 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Geometric alignment and synthetic-to-real adaptation · 94 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
adapt IC13 training images with GA-DANEAST · ICDAR2015F-score: 53.3 -> 64.4Improved
adapt IC13 training images with GA-DANEAST · MSRA-TD500F-score: 46.8 -> 58.5Improved
generate 10 adapted images per source image instead of 1 (multi-modal spatial learning) (ablation)EAST · ICDAR2015F-score: 64.4 -> 69.4Improved
generate 10 adapted images per source image instead of 1 (multi-modal spatial learning) (ablation)EAST · MSRA-TD500F-score: 58.5 -> 62.1Improved
add 10x adapted IC13 images to target training dataEAST · ICDAR2015F-score: 80.0 -> 83.5Improved
add 10x adapted IC13 images to target training dataEAST · MSRA-TD500F-score: 70.3 -> 75.5Improved
use GA-DAN instead of CycleGAN for adaptationEAST · ICDAR2015F-score: 57.2 -> 64.4Improved
use GA-DAN instead of ST-GAN + CycleGAN concatenationEAST · ICDAR2015F-score: 60.8 -> 64.4Improved
add disentangled cycle-consistency loss (vs normal cycle loss) (ablation)GA-DAN + MORAN · COMB-CUTEaccuracy: 32.6 -> 38.2Improved
add disentangled cycle-consistency loss (vs normal cycle loss) (ablation)GA-DAN + MORAN · COMB-SVTPaccuracy: 44.9 -> 47.1Improved
add appearance adaptation to spatial-only output (ablation)GA-DAN + MORAN · COMB-CUTEaccuracy: 36.1 -> 38.2Improved
generate 10 adapted images per source image instead of 1 (ablation)GA-DAN + MORAN · COMB-CUTEaccuracy: 38.2 -> 43.1Improved

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