Research map / Image generation / GANs
Edge Guided GANs With Multi-Scale Contrastive Learning for Semantic Image Synthesis
Hao Tang, Guolei Sun, Nicu Sebe, Luc Van Gool
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add edge generator G_e (ablation) | E+G_i baseline · Cityscapes | mIoU: 58.6 -> 60.2 | Improved |
| add edge generator G_e (ablation) | E+G_i baseline · ADE20K | mIoU: 36.9 -> 38.7 | Improved |
| add edge generator G_e (ablation) | E+G_i baseline · COCO-Stuff | mIoU: 36.8 -> 37.5 | Improved |
| add attention guided edge transfer G_t (ablation) | E+G_i+G_e · ADE20K | mIoU: 38.7 -> 40.6 | Improved |
| add semantic preserving module G_s (ablation) | E+G_i+G_e+G_t · Cityscapes | mIoU: 61.5 -> 64.5 | Improved |
| add G_l component (ablation) | B4 model · ADE20K | mIoU: 42.0 -> 45.8 | Improved |
| add G_c component (class-specific generation) (ablation) | B5 model · Cityscapes | FID: 52.2 -> 44.5 | Improved |
| add G_c component (ablation) | B5 model · ADE20K | mIoU: 45.8 -> 50.6 | Improved |
| add first multi-scale contrastive loss term (ablation) | ECGAN (B6) · ADE20K | mIoU: 50.6 -> 51.6 | Improved |
| add second cross-scale loss term (ECGAN++) (ablation) | B7 · ADE20K | mIoU: 51.6 -> 52.7 | Improved |
| full ECGAN++ over ECGAN | ECGAN · COCO-Stuff | FID: 15.7 -> 14.9 | Improved |
| ECGAN++ vs OASIS on small objects | OASIS · Cityscapes small objects | mIoU: 26.0 -> 29.4 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- OASIS — OASIS: Only Adversarial Supervision for Semantic Image Synthesis (IJCV 2022) · better on 8 of 8 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Edge Guided GANs with Contrastive Learning for Semantic Image Synthesis (ICLR 2023 · similarity 0.98)
- Learning to Predict Layout-to-image Conditional Convolutions for Semantic Image Synthesis (NeurIPS 2019 · similarity 0.91)
- OASIS: Only Adversarial Supervision for Semantic Image Synthesis (IJCV 2022 · similarity 0.90)
- Linear Semantics in Generative Adversarial Networks (CVPR 2021 · similarity 0.90)
- You Only Need Adversarial Supervision for Semantic Image Synthesis (ICLR 2021 · similarity 0.90)
- Unlocking Pre-trained Image Backbones for Semantic Image Synthesis (CVPR 2024 · similarity 0.90)
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.
