Research map / Image generation / GANs
CAFE-GAN: Arbitrary Face Attribute Editing with Complementary Attention Feature
Jeong gi Kwak, David K. Han, Hanseok Ko
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 complementary matching loss (ablation) | CAFE-GAN · CelebA 128x128 attribute editing | average attribute classification accuracy: 81.02 -> 85.64 | Improved |
| add complementary attention branch (CAB) in discriminator (ablation) | CAFE-GAN · CelebA 128x128 attribute editing | average attribute classification accuracy: 70.66 -> 85.64 | Improved |
| add complementary matching loss (ablation) | CAFE-GAN · CelebA 128x128 attribute editing | accuracy on Bald: 61.68 -> 79.03 | Improved |
| add complementary attention branch (ablation) | CAFE-GAN · CelebA 128x128 attribute editing | accuracy on Bald: 32.43 -> 79.03 | Improved |
| add complementary attention branch (ablation) | CAFE-GAN · CelebA 128x128 attribute editing | accuracy on Aged: 65.95 -> 88.61 | Improved |
| add complementary matching loss (ablation) | CAFE-GAN · CelebA 128x128 attribute editing | accuracy on Open mouth: 97.56 -> 97.15 | Hurt |
| add complementary matching loss (ablation) | CAFE-GAN · CelebA 128x128 attribute editing | accuracy on Gender: 85.93 -> 95.22 | Improved |
| add complementary attention branch (ablation) | CAFE-GAN · CelebA 128x128 attribute editing | accuracy on Blond hair: 70.41 -> 88.14 | Improved |
| CAFE-GAN vs STGAN | STGAN · CelebA 128x128 attribute editing | average attribute classification accuracy: 81.54 -> 85.64 | Improved |
| CAFE-GAN vs AttGAN | AttGAN · CelebA 128x128 attribute editing | average attribute classification accuracy: 63.72 -> 85.64 | Improved |
| CAFE-GAN vs STGAN on Bald | STGAN · CelebA 128x128 attribute editing | accuracy on Bald: 59.76 -> 79.03 | Improved |
| CAFE-GAN vs STGAN on Mustache | STGAN · CelebA 128x128 attribute editing | accuracy on Mustache: 42.10 -> 40.13 | Hurt |
Closest papers by meaning
- L2M-GAN: Learning To Manipulate Latent Space Semantics for Facial Attribute Editing (CVPR 2021 · similarity 0.89)
- Generative Adversarial Network with Spatial Attention for Face Attribute Editing (ECCV 2018 · similarity 0.89)
- SSCGAN: Facial Attribute Editing via Style Skip Connections (ECCV 2020 · similarity 0.88)
- SDGAN: Disentangling Semantic Manipulation for Facial Attribute Editing (AAAI 2024 · similarity 0.88)
- Mask-Guided Portrait Editing With Conditional GANs (CVPR 2019 · similarity 0.88)
- Image-to-Image Translation With Disentangled Latent Vectors for Face Editing (TPAMI 2023 · 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.
