AI Paper Atlas

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

Domain Randomization and Pyramid Consistency: Simulation-to-Real Generalization Without Accessing Target Domain Data

Xiangyu Yue, Yang Zhang, Sicheng Zhao, Alberto Sangiovanni-Vincentelli, Kurt Keutzer, Boqing Gong

ICCV 2019 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Domain-generalized segmentation · 400 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 domain randomization with auxiliary stylized domains (ablation)FCN8s-VGG16 · GTA -> CityscapesmIoU: 30.04 -> 34.64Improved
add domain randomization (ablation)FCN8s-VGG16 · GTA -> BDDSmIoU: 24.59 -> 30.14Improved
add domain randomization (ablation)FCN8s-VGG16 · GTA -> MapillarymIoU: 26.63 -> 31.64Improved
add pyramid consistency across domains on top of DR (ablation)FCN8s-VGG16 · GTA -> CityscapesmIoU: 34.64 -> 35.47Improved
add pyramid consistency within an image on top of DR (ablation)FCN8s-VGG16 · GTA -> CityscapesmIoU: 34.64 -> 35.12Improved
add both consistency losses on top of DR (ablation)FCN8s-VGG16 · GTA -> CityscapesmIoU: 34.64 -> 36.11Improved
add both consistency losses on top of DR (ablation)FCN8s-VGG16 · GTA -> MapillarymIoU: 31.64 -> 32.25Improved
increase number of auxiliary domains (ablation)FCN8s-VGG16 · GTA -> Cityscapes/BDDS/MapillarymIoU: accuracy increases then saturates (Figure 4, no numbers)Mixed
apply full method with ResNet-101 backboneFCN-ResNet-101 · GTA -> CityscapesmIoU: 33.56 -> 42.53Improved
apply full method versus IBN-NetResNet-50 · GTA -> CityscapesmIoU: 29.64 -> 37.42Improved
use validation on target in the generalization methodOurs · GTA -> Cityscapes (VGG-16)mIoU: 36.1 -> 38.6Improved
apply full method on SYNTHIA sourceFCN-VGG-16 · SYNTHIA -> Cityscapes/BDDS/MapillaryGperf mIoU: 25.34 -> 32.41Improved

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