Research map / Image generation / Editing and style transfer
Unpaired Image-to-Image Translation via Latent Energy Transport
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 latent energy transport with a pretrained ALAE instead of StarGAN v2 | LETIT · CelebA-HQ | FID: 13.7 -> 12.5 | Helped |
| Use latent energy transport with a pretrained ALAE instead of StarGAN v2 | LETIT · AFHQ | FID: 16.2 -> 15.9 | Helped |
| Use LETIT instead of StarGAN v2 | LETIT · CelebA-HQ | Faith: 2.2 -> 51.7 | Helped |
| Use LETIT instead of StarGAN v2 | LETIT · AFHQ | Faith: 38.4 -> 61.6 | Helped |
| Use LETIT instead of CUT | LETIT with VQ-VAE-2 · AFHQ cat → dog | FID: 76.2 -> 45.2 | Helped |
| Use LETIT instead of CUT | LETIT with VQ-VAE-2 · AFHQ cat → dog | Mem(GB): 3.03 -> 1.24 | Helped |
| Use LETIT instead of CF-EBM | LETIT with VQ-VAE-2 · photo → vangogh | KID: 4.25 -> 4.42 | Hurt |
| Use LETIT instead of CF-EBM | LETIT with VQ-VAE-2 · photo → vangogh | Days: 0.6 -> 0.1 | Helped |
| Pretrain ALAE on FFHQ rather than CelebA-HQ (ablation) | LETIT (Ours-T) · CelebA-HQ | Quality: 35.4 -> 39.0 | Helped |
| Pretrain ALAE on FFHQ rather than CelebA-HQ (ablation) | LETIT (Ours-T) · CelebA-HQ | Faith: 51.7 -> 46.1 | Hurt |
| Use latent transport with a VQ-VAE-2 pretrained on AFHQ rather than the task-specific dataset (ablation) | LETIT with VQ-VAE-2 · AFHQ cat → dog | KID: 6.20 -> 6.01 | Helped |
| Use latent transport with a VQ-VAE-2 pretrained on AFHQ rather than the task-specific dataset (ablation) | LETIT with VQ-VAE-2 · AFHQ dog → cat | KID: 9.21 -> 7.45 | Helped |
Closest papers by meaning
- StEP: Style-Based Encoder Pre-Training for Multi-Modal Image Synthesis (CVPR 2021 · similarity 0.89)
- Rethinking the Paradigm of Content Constraints in Unpaired Image-to-Image Translation (AAAI 2024 · similarity 0.88)
- Latent Filter Scaling for Multimodal Unsupervised Image-To-Image Translation (CVPR 2019 · similarity 0.88)
- EGSDE: Unpaired Image-to-Image Translation via Energy-Guided Stochastic Differential Equations (NeurIPS 2022 · similarity 0.88)
- Unsupervised Image-to-Image Translation With Generative Prior (CVPR 2022 · similarity 0.88)
- Unpaired Image-to-Image Translation via Neural Schrödinger Bridge (ICLR 2024 · similarity 0.86)
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.