Research map / Image generation / GANs
Exploiting Spatial Dimensions of Latent in GAN for Real-Time Image Editing
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 |
|---|---|---|---|
| Add adversarial loss (ablation) | StyleMapGAN · CelebA-HQ 256×256, 8×8 stylemap | FID_lerp: 12.99 -> 9.97 | Helped |
| Add domain-guided loss (ablation) | StyleMapGAN · CelebA-HQ 256×256, 8×8 stylemap | FID_lerp: 16.84 -> 9.97 | Helped |
| Add latent reconstruction loss (ablation) | StyleMapGAN · CelebA-HQ 256×256, 8×8 stylemap | FID_lerp: 10.08 -> 9.97 | Helped |
| Add image reconstruction loss (ablation) | StyleMapGAN · CelebA-HQ 256×256, 8×8 stylemap | FID_lerp: 10.29 -> 9.97 | Helped |
| Add perceptual loss (ablation) | StyleMapGAN · CelebA-HQ 256×256, 8×8 stylemap | FID_lerp: 13.33 -> 9.97 | Helped |
| Add R1 regularization (ablation) | StyleMapGAN · CelebA-HQ 256×256, 8×8 stylemap | FID_lerp: 14.56 -> 9.97 | Helped |
| Use joint generator-encoder learning instead of sequential training (ablation) | StyleMapGAN-Light · FFHQ 1024×1024 | FID_lerp: 201.18 -> 13.52 | Helped |
| Use an 8×8 rather than 4×4 stylemap (ablation) | StyleMapGAN · CelebA-HQ | FID_lerp: 10.72 -> 9.97 | Helped |
| Use a 16×16 rather than 8×8 stylemap (ablation) | StyleMapGAN · CelebA-HQ | FID_lerp: 9.97 -> 11.05 | Hurt |
| Use a 32×32 rather than 8×8 stylemap (ablation) | StyleMapGAN · CelebA-HQ | FID_lerp: 9.97 -> 12.10 | Hurt |
Closest papers by meaning
- Navigating the GAN Parameter Space for Semantic Image Editing (CVPR 2021 · similarity 0.92)
- Editing in Style: Uncovering the Local Semantics of GANs (CVPR 2020 · similarity 0.91)
- Enjoy Your Editing: Controllable GANs for Image Editing via Latent Space Navigation (ICLR 2021 · similarity 0.90)
- E$^2$GAN: Efficient Training of Efficient GANs for Image-to-Image Translation (ICML 2024 · similarity 0.90)
- WeditGAN: Few-Shot Image Generation via Latent Space Relocation (AAAI 2024 · similarity 0.88)
- Reference-Based 3D-Aware Image Editing with Triplanes (CVPR 2025 · similarity 0.88)
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.