Research map / Image generation / GANs
Mask-Embedded Discriminator With Region-Based Semantic Regularization for Semi-Supervised Class-Conditional Image Synthesis
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use MED-GAN instead of R3-CGAN | Semi-supervised class-conditional synthesis · CIFAR-10 (4k) | FID: 20.34 -> 5.76 | Helped |
| Use MED-GAN instead of R3-CGAN | Semi-supervised class-conditional synthesis · CIFAR-100 (10k) | FID: 26.29 -> 8.06 | Helped |
| Use MED-GAN instead of R3-CGAN | Semi-supervised class-conditional synthesis · FS-100 (2k) | FID: 25.28 -> 14.42 | Helped |
| Use MED-GAN instead of R3-CGAN | Semi-supervised class-conditional synthesis · CUB-200 (2.8k) | FID: 88.62 -> 16.90 | Helped |
| Use MED-GAN instead of the strongest listed baseline | Semi-supervised classifier · CIFAR-10 | test error rate (%): 6.44 -> 6.02 | Helped |
| Use MED-GAN instead of R3-CGAN | Semi-supervised classifier · CIFAR-100 | test error rate (%): 32.66 -> 30.67 | Helped |
| Use MED-GAN instead of R3-CGAN | Semi-supervised classifier · FS-100 | test error rate (%): 6.96 -> 5.71 | Helped |
| Add mask embedding to the baseline discriminator (ablation) | Semi-supervised GAN · CUB-200 | FID: 35.04 -> 23.73 | Helped |
| Add region-based regularization after mask embedding (ablation) | Semi-supervised GAN · CUB-200 | FID: 23.73 -> 17.94 | Helped |
| Add mask embedding to the baseline discriminator (ablation) | Fully supervised BigGAN · CIFAR-10 | FID: 9.06 -> 7.43 | Helped |
| Add region-based regularization after mask embedding (ablation) | Fully supervised BigGAN · CIFAR-10 | FID: 7.43 -> 5.01 | Helped |
| Add recognition loss and contrastive loss to the baseline (ablation) | Synthetic-data baseline GAN · 2D-ring synthetic dataset | # Modes: 5.3±1.1 -> 8.0±0.0 | Helped |
Closest papers by meaning
- Regularizing Discriminative Capability of CGANs for Semi-Supervised Generative Learning (CVPR 2020 · similarity 0.91)
- Exploring Intra-Class Variation Factors With Learnable Cluster Prompts for Semi-Supervised Image Synthesis (CVPR 2023 · similarity 0.89)
- OSSGAN: Open-Set Semi-Supervised Image Generation (CVPR 2022 · similarity 0.88)
- Conditional Image Synthesis with Auxiliary Classifier GANs (ICML 2017 · similarity 0.88)
- Improving GAN with Neighbors Embedding and Gradient Matching (AAAI 2019 · similarity 0.87)
- Unlocking Pre-trained Image Backbones for Semantic Image Synthesis (CVPR 2024 · similarity 0.85)
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.