Research map / Image generation / Editing and style transfer
Swapping Autoencoder for Deep Image Manipulation
Read the paperCheck 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 the encoder-based Swapping Autoencoder instead of optimization-based image embedding | Image embedding for manipulation · Test-image embedding | Runtime: 0.113 vs 990 | Helped |
| Use the Swapping Autoencoder rather than StyleGAN2 | Image hybrid generation · Average across Church, FFHQ, and Waterfall | AMT Fooling Rate: 31.0$\pm$1.4 vs 24.4$\pm$1.4 | Helped |
| Use the Swapping Autoencoder rather than WCT$^{2}$ | Image hybrid generation · Average across Church, FFHQ, and Waterfall | FID: 54.36 vs 36.85 | Hurt |
| Increase the maximum crop size seen by the co-occurrence discriminator (ablation) | Swapping Autoencoder · Image swapping | Self-similarity distance | Mixed |
| Increase the maximum crop size seen by the co-occurrence discriminator (ablation) | Swapping Autoencoder · Image swapping | SIFID | Mixed |
| Interpolate between texture codes rather than fully swap them (ablation) | Swapping Autoencoder · Image swapping | SIFID | No clear effect |
| Interpolate between texture codes rather than fully swap them (ablation) | Swapping Autoencoder · Image swapping | Self-similarity distance | No clear effect |
Closest papers by meaning
- Deep Automodulators (NeurIPS 2020 · similarity 0.88)
- Exploring the Effectiveness of Mask-Guided Feature Modulation as a Mechanism for Localized Style Editing of Real Images (Student Abstract) (AAAI 2023 · similarity 0.86)
- PetsGAN: Rethinking Priors for Single Image Generation (AAAI 2022 · similarity 0.86)
- Learning to Manipulate Artistic Images (AAAI 2024 · similarity 0.85)
- Bringing NeRFs to the Latent Space: Inverse Graphics Autoencoder (ICLR 2025 · similarity 0.85)
- EditInfinity: Image Editing with Binary-Quantized Generative Models (NeurIPS 2025 · 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.