AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

DGInStyle: Domain-Generalizable Semantic Segmentation with Image Diffusion Models and Stylized Semantic Control

Yuru Jia, Lukas Hoyer, Shengyu Huang, Tianfu Wang, Luc Van Gool, Konrad Schindler, Anton Obukhov

ECCV 2024 · Image generation › Fast diffusion sampling · approach: Diffusion models as perception tools · 48 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 DGInStyle generated data to DG methods trained on GTAIBN-Net, RobustNet, HRDA, Color-Aug, DAFormer · GTA to 5 real datasetsmIoU (Avg5): +2.5 to +7.2 mIoUImproved
add DGInStyle data when source is CityscapesColor-Aug, DAFormer, HRDA · Cityscapes to BDD, MV, ACDC, DZmIoU (average): +2.1, +1.5, +0.7Improved
add Multi-Resolution Latent Fusion (ablation)DGInStyle with DAFormer · GTA to 5 real datasetsmIoU (Avg5): 43.31 -> 44.27Improved
add Style Swap (ablation)DGInStyle with DAFormer · GTA to 5 real datasetsmIoU (Avg5): 44.27 -> 45.84Improved
add Style Prompts (ablation)DGInStyle with DAFormer · GTA to 5 real datasetsmIoU (Avg5): 45.84 -> 46.16Improved
add Rare Class Generation (ablation)DGInStyle with DAFormer · GTA to 5 real datasetsmIoU (Avg5): 46.16 -> 46.47Improved
add Controlled Tiled MultiDiffusion within MRLF (ablation)DGInStyle with DAFormer · GTA to 5 real datasetsmIoU (Avg5): 45.19 -> 45.60Improved
add Latent Inpainting Diffusion to CTMD (ablation)DGInStyle with DAFormer · GTA to 5 real datasetsmIoU (Avg5): 45.60 -> 46.47Improved
add Style Swap and Style Prompting to generation (ablation)DGInStyle generated data · FID to 5 real datasets (average)FID: 136.68 -> 127.37Improved
increase generated images from 6000 to 8000 (ablation)DGInStyle with DAFormer · GTA to 5 real datasetsmIoU (Avg5): 46.47 -> 46.39No clear effect
add DGInStyle data to HRDA with ResNet-101HRDA ResNet-101 · GTA to Dark ZurichmIoU: 7.80 -> 16.16Improved

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