Research map / Image generation / GANs
SphericGAN: Semi-Supervised Hyper-Spherical Generative Adversarial Networks for Fine-Grained 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 SphericGAN instead of SSC-GAN | class-conditional fine-grained image generation · CUB-200 | FID: 20.03 -> 18.87 | Helped |
| use SphericGAN instead of BigGAN | class-conditional fine-grained image generation · FaceScrub-100 | FID: 15.51 -> 18.84 | Hurt |
| use SphericGAN instead of BigGAN | class-conditional fine-grained image generation · Stanford-Cars | FID: 32.45 -> 35.69 | Hurt |
| add latent clustering (ablation) | SphericGAN baseline · CUB-200 | FID: 79.12 -> 50.20 | Helped |
| add latent clustering (ablation) | SphericGAN baseline · CUB-200 | IS: 4.34±.05 -> 4.39±.03 | Helped |
| add semantic verification of synthesized data (ablation) | SphericGAN with latent clustering · CUB-200 | RA: 9.88 -> 92.04 | Helped |
| add semantic verification of synthesized data (ablation) | SphericGAN with latent clustering · CUB-200 | cFID: 165.23 -> 115.35 | Helped |
| add real-fake cluster alignment (ablation) | SphericGAN with latent clustering and semantic verification · CUB-200 | FID: 30.15 -> 18.87 | Helped |
| add real-fake cluster alignment (ablation) | SphericGAN with latent clustering and semantic verification · CUB-200 | cFID: 115.35 -> 103.35 | Helped |
| add real-fake cluster alignment (ablation) | SphericGAN with latent clustering and semantic verification · CUB-200 | RA: 92.04 -> 98.46 | Helped |
Closest papers by meaning
- Large Scale GAN Training for High Fidelity Natural Image Synthesis (ICLR 2019 · similarity 0.89)
- DeLiGAN : Generative Adversarial Networks for Diverse and Limited Data (CVPR 2017 · similarity 0.88)
- A Geometric Analysis of Deep Generative Image Models and Its Applications (ICLR 2021 · similarity 0.88)
- Regularizing Discriminative Capability of CGANs for Semi-Supervised Generative Learning (CVPR 2020 · similarity 0.88)
- CircleGAN: Generative Adversarial Learning across Spherical Circles (NeurIPS 2020 · similarity 0.87)
- High-Fidelity Image Generation With Fewer Labels (ICML 2019 · similarity 0.87)
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.