Research map / Image generation / Editing and style transfer
Latent Filter Scaling for Multimodal Unsupervised Image-To-Image Translation
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-code-mapped scalars to modulate convolutional feature maps instead of injecting the latent code as image or feature data | Latent Filter Scaling · Edges to shoes | LPIPS diversity: 0.104 -> 0.109 | Helped |
| Use latent-code-mapped scalars to modulate convolutional feature maps instead of injecting the latent code as image or feature data | Latent Filter Scaling · Edges to shoes | AMT preference: 45.1% | Helped |
| Use latent-code-mapped scalars to modulate convolutional feature maps instead of injecting the latent code as image or feature data | Latent Filter Scaling · Edges to shoes | AMT preference: 57.2% | Hurt |
| Use tanh activation in the final scalar-mapping layer (ablation) | Latent Filter Scaling · Image-to-image translation | Image quality | Helped |
| Use tanh activation in the final scalar-mapping layer (ablation) | Latent Filter Scaling · Image-to-image translation | Output variability | Hurt |
| Replace tanh with LRELU or a linear mapping (ablation) | Latent Filter Scaling · Image-to-image translation | Output variability | Helped |
| Replace tanh with LRELU or a linear mapping (ablation) | Latent Filter Scaling · Image-to-image translation | Image quality | Hurt |
| Add bias to the fully connected scalar-mapping network (ablation) | Latent Filter Scaling · Image-to-image translation | Image quality | Helped |
| Add bias to the fully connected scalar-mapping network (ablation) | Latent Filter Scaling · Image-to-image translation | Output variability | Hurt |
Closest papers by meaning
- Toward Multimodal Image-to-Image Translation (NeurIPS 2017 · similarity 0.91)
- Unsupervised Image-to-Image Translation Networks (NeurIPS 2017 · similarity 0.90)
- Unsupervised Image-to-Image Translation With Generative Prior (CVPR 2022 · similarity 0.90)
- TraVeLGAN: Image-To-Image Translation by Transformation Vector Learning (CVPR 2019 · similarity 0.90)
- Self-Supervised Dense Consistency Regularization for Image-to-Image Translation (CVPR 2022 · similarity 0.89)
- Rethinking the Paradigm of Content Constraints in Unpaired Image-to-Image Translation (AAAI 2024 · 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.