Research map / Image generation / Editing and style transfer
Mask-Guided Portrait Editing With Conditional GANs
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 the background fusing sub-network to local component embeddings (ablation) | Mask-Guided Portrait Editing · Helen-derived generated faces | FID: 11.26 -> 8.92 | Helped |
| Replace global embedding with local component embeddings (ablation) | Mask-Guided Portrait Editing with direct background pasting · Helen-derived generated faces | FID: 11.02 -> 11.26 | No clear effect |
| Add generated faces to the Helen training set with standard data augmentation | Helen face parsing · Helen face parsing test set | accuracy: 0.863 -> 0.871 | Helped |
| Add standard data augmentation | Helen face parsing · Helen face parsing test set | accuracy: 0.728 -> 0.863 | Helped |
| Add the face parsing loss to constrain generated masks (ablation) | Mask-Guided Portrait Editing · Generated faces evaluated against target masks | average per-pixel accuracy: 0.946 -> 0.977 | Helped |
| Use local component embeddings instead of a global embedding (ablation) | Mask-Guided Portrait Editing · Portrait synthesis and editing | component detail and controllability | Helped |
| Add the background fusing sub-network instead of directly pasting the background (ablation) | Mask-Guided Portrait Editing · Portrait synthesis | boundary artifacts | Helped |
Closest papers by meaning
- Interpreting the Latent Space of GANs for Semantic Face Editing (CVPR 2020 · similarity 0.90)
- Neural Face Editing With Intrinsic Image Disentangling (CVPR 2017 · similarity 0.90)
- MaskGAN: Towards Diverse and Interactive Facial Image Manipulation (CVPR 2020 · similarity 0.90)
- FaceController: Controllable Attribute Editing for Face in the Wild (AAAI 2021 · similarity 0.89)
- MOST-GAN: 3D Morphable StyleGAN for Disentangled Face Image Manipulation (AAAI 2022 · similarity 0.89)
- In-N-Out: Faithful 3D GAN Inversion with Volumetric Decomposition for Face Editing (CVPR 2024 · 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.