paper-with-me

Synthetic-to-Real Translation 벤치마크

Synthetic-to-Real Translation on SYNTHIA-to-Cityscapes

114개 결과 · ⬇ CSV · JSON

MIoU (16 classes)

44.5 50.7 56.9 63.1 69.3 2019-10 2026-09 CAG-UDA — 44.5 (2019-10-29) CAG-UDA — 44.5 (2019-10-29) CAG-UDA — 44.5 (2019-10-29) MRNet+Rectifying Label(ResNet-101) — 47.9 (2020-03-08) MRNet+Rectifying Label(ResNet-101) — 47.9 (2020-03-08) MRNet+Rectifying Label(ResNet-101) — 47.9 (2020-03-08) DACS(ResNet-101) — 48.34 (2020-07-17) FADA(ResNet-101) — 45.2 (2020-07-17) DACS(ResNet-101) — 48.34 (2020-07-17) FADA(ResNet-101) — 45.2 (2020-07-17) DACS(ResNet-101) — 48.34 (2020-07-17) FADA(ResNet-101) — 45.2 (2020-07-17) IAST(ResNet-101) — 49.8 (2020-08-27) IAST(ResNet-101) — 49.8 (2020-08-27) IAST(ResNet-101) — 49.8 (2020-08-27) PLCA (ResNet-101) — 46.8 (2020-10-31) PLCA (ResNet-101) — 46.8 (2020-10-31) PLCA (ResNet-101) — 46.8 (2020-10-31) ProDA(ResNet-101) — 55.5 (2021-01-26) ProDA(ResNet-101) — 55.5 (2021-01-26) ProDA(ResNet-101) — 55.5 (2021-01-26) MetaCorrection(ResNet-101) — 45.1 (2021-03-09) MetaCorrection(ResNet-101) — 45.1 (2021-03-09) MetaCorrection(ResNet-101) — 45.1 (2021-03-09) Coarse-to-Fine(ResNet-101) — 48.2 (2021-03-24) Coarse-to-Fine(ResNet-101) — 48.2 (2021-03-24) Coarse-to-Fine(ResNet-101) — 48.2 (2021-03-24) Uncertainty + Adaboost (ResNet-101) — 50.4 (2021-03-29) MRNet + Adaboost (ResNet-101) — 45.9 (2021-03-29) Uncertainty + Adaboost (ResNet-101) — 50.4 (2021-03-29) MRNet + Adaboost (ResNet-101) — 45.9 (2021-03-29) Uncertainty + Adaboost (ResNet-101) — 50.4 (2021-03-29) MRNet + Adaboost (ResNet-101) — 45.9 (2021-03-29) CorDA(ResNet-101) — 55.0 (2021-04-28) CorDA(ResNet-101) — 55.0 (2021-04-28) CorDA(ResNet-101) — 55.0 (2021-04-28) SAC(ResNet-101) — 52.6 (2021-04-30) SAC(ResNet-101) — 52.6 (2021-04-30) SAC(ResNet-101) — 52.6 (2021-04-30) CLST(ResNet-101) — 49.8 (2021-05-05) CLST(ResNet-101) — 49.8 (2021-05-05) CLST(ResNet-101) — 49.8 (2021-05-05) PixMatch(ResNet-101) — 46.1 (2021-05-17) PixMatch(ResNet-101) — 46.1 (2021-05-17) PixMatch(ResNet-101) — 46.1 (2021-05-17) ITEN — 46.45 (2021-07-13) ITEN — 46.45 (2021-07-13) ITEN — 46.45 (2021-07-13) DSP(ResNet-101) — 51.0 (2021-07-20) DSP(ResNet-101) — 51.0 (2021-07-20) DSP(ResNet-101) — 51.0 (2021-07-20) ProDA+CRA — 56.9 (2021-09-14) ProDA+CRA — 56.9 (2021-09-14) ProDA+CRA — 56.9 (2021-09-14) DAFormer — 60.9 (2021-11-29) DAFormer — 60.9 (2021-11-29) DAFormer — 60.9 (2021-11-29) TransDA-B — 59.3 (2022-03-15) TransDA-B — 59.3 (2022-03-15) TransDA-B — 59.3 (2022-03-15) SePiCo — 64.3 (2022-04-19) SePiCo (ResNet-101) — 58.1 (2022-04-19) SePiCo — 64.3 (2022-04-19) SePiCo (ResNet-101) — 58.1 (2022-04-19) SePiCo — 64.3 (2022-04-19) SePiCo (ResNet-101) — 58.1 (2022-04-19) DAFormer + ProCST — 61.6 (2022-04-25) DAFormer + ProCST — 61.6 (2022-04-25) DAFormer + ProCST — 61.6 (2022-04-25) HRDA — 65.8 (2022-04-27) HRDA — 65.8 (2022-04-27) HRDA — 65.8 (2022-04-27) EHTDI* — 61.3 (2022-08-12) EHTDI — 57.8 (2022-08-12) EHTDI* — 61.3 (2022-08-12) EHTDI — 57.8 (2022-08-12) EHTDI* — 61.3 (2022-08-12) EHTDI — 57.8 (2022-08-12) CLUDA+HRDA — 67.2 (2022-08-27) CLUDA+HRDA — 67.2 (2022-08-27) CLUDA+HRDA — 67.2 (2022-08-27) G2L — 56.8 (2022-09-07) G2L — 56.8 (2022-09-07) G2L — 56.8 (2022-09-07) HRDA+PiPa — 68.2 (2022-11-14) HRDA+PiPa — 68.2 (2022-11-14) HRDA+PiPa — 68.2 (2022-11-14) MIC — 67.3 (2022-12-02) MIC — 67.3 (2022-12-02) MIC — 67.3 (2022-12-02) Sepico + HIAST — 59.6 (2023-02-14) AdaptSeg + HIAST — 53.5 (2023-02-14) Sepico + HIAST — 59.6 (2023-02-14) AdaptSeg + HIAST — 53.5 (2023-02-14) Sepico + HIAST — 59.6 (2023-02-14) AdaptSeg + HIAST — 53.5 (2023-02-14) DCF — 69.3 (2023-11-21) DCF — 69.3 (2023-11-21) DCF — 69.3 (2023-11-21) CAG-UDA — 44.5 (2019-10-29) MRNet+Rectifying Label(ResNet-101) — 47.9 (2020-03-08) DACS(ResNet-101) — 48.34 (2020-07-17) IAST(ResNet-101) — 49.8 (2020-08-27) ProDA(ResNet-101) — 55.5 (2021-01-26) ProDA+CRA — 56.9 (2021-09-14) DAFormer — 60.9 (2021-11-29) SePiCo — 64.3 (2022-04-19) HRDA — 65.8 (2022-04-27) CLUDA+HRDA — 67.2 (2022-08-27) HRDA+PiPa — 68.2 (2022-11-14) DCF — 69.3 (2023-11-21)
RankModel MIoU (16 classes)MIoU (13 classes) Extra Training Data PaperCodeYear
1 DCF 69.375.9 ✓ Transferring to Real-World Layouts: A Depth-aware Framework for Scene Adaptation chen742/PiPa · chen742/DCF 2023
2 HRDA+PiPa 68.274.8 PiPa: Pixel- and Patch-wise Self-supervised Learning for Domain Adaptative Semantic Segmentation chen742/PiPa 2022
3 MIC 67.374.0 MIC: Masked Image Consistency for Context-Enhanced Domain Adaptation lhoyer/mic 2022
4 CLUDA+HRDA 67.2– CLUDA : Contrastive Learning in Unsupervised Domain Adaptation for Semantic Segmentation user0407/CLUDA 2022
5 HRDA 65.872.4 HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentation lhoyer/hrda 2022
6 SePiCo 64.371.4 SePiCo: Semantic-Guided Pixel Contrast for Domain Adaptive Semantic Segmentation bit-da/sepico 2022
7 DAFormer + ProCST 61.6– ✓ ProCST: Boosting Semantic Segmentation Using Progressive Cyclic Style-Transfer shahaf1313/procst 2022
8 EHTDI* 61.369.2 Exploring High-quality Target Domain Information for Unsupervised Domain Adaptive Semantic Segmentation ljjcoder/ehtdi 2022
9 DAFormer 60.967.4 DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation lhoyer/DAFormer · dbash/visda2022-org · kw01sg/crda 2021
10 Sepico + HIAST 59.668.1 Hard-aware Instance Adaptive Self-training for Unsupervised Cross-domain Semantic Segmentation bupt-ai-cz/hiast 2023
11 TransDA-B 59.366.3 Smoothing Matters: Momentum Transformer for Domain Adaptive Semantic Segmentation alpc91/transda 2022
12 SePiCo (ResNet-101) 58.166.5 SePiCo: Semantic-Guided Pixel Contrast for Domain Adaptive Semantic Segmentation bit-da/sepico 2022
13 EHTDI 57.864.6 Exploring High-quality Target Domain Information for Unsupervised Domain Adaptive Semantic Segmentation ljjcoder/ehtdi 2022
14 ProDA+CRA 56.963.7 Cross-Region Domain Adaptation for Class-level Alignment 2021
15 G2L 56.864.4 G2L: A Global to Local Alignment Method for Unsupervised Domain Adaptive Semantic Segmentation 2022
16 ProDA(ResNet-101) 55.562.0 Prototypical Pseudo Label Denoising and Target Structure Learning for Domain Adaptive Semantic Segmentation microsoft/ProDA · E-DEEP/PapersReview 2021
17 CorDA(ResNet-101) 55.062.8 ✓ Domain Adaptive Semantic Segmentation with Self-Supervised Depth Estimation qinenergy/corda 2021
18 AdaptSeg + HIAST 53.560.3 Hard-aware Instance Adaptive Self-training for Unsupervised Cross-domain Semantic Segmentation bupt-ai-cz/hiast 2023
19 SAC(ResNet-101) 52.659.3 Self-supervised Augmentation Consistency for Adapting Semantic Segmentation visinf/da-sac 2021
20 DSP(ResNet-101) 51.059.9 DSP: Dual Soft-Paste for Unsupervised Domain Adaptive Semantic Segmentation GaoLii/DSP 2021
1–20 / 114 다음 → 페이지당 10 20 50 100