Research map / Image generation / Text-to-image diffusion
ManiCLIP: Multi-attribute Face Manipulation from Text
Hao Wang, Guosheng Lin, Ana García del Molino, Anran Wang, Jiashi Feng, Zhiqi Shen
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 decoupling training (group sampling) to the baseline (ablation) | ManiCLIP · Multi-Modal-CelebA-HQ test (5,000) | FID: 40.70 -> 34.60 | Improved |
| add entropy loss to the baseline (ablation) | ManiCLIP · Multi-Modal-CelebA-HQ test (5,000) | FID: 40.70 -> 38.62 | Improved |
| add entropy loss to decoupling training (ablation) | ManiCLIP · Multi-Modal-CelebA-HQ test (5,000) | FID: 34.60 -> 31.57 | Improved |
| add decoupling training to entropy-loss model (ablation) | ManiCLIP · Multi-Modal-CelebA-HQ test (5,000) | FID: 38.62 -> 31.57 | Improved |
| add decoupling training to the baseline (ablation) | ManiCLIP · Multi-Modal-CelebA-HQ test (5,000) | ID similarity: 69.88 -> 74.41 | Improved |
| add entropy loss to the baseline (ablation) | ManiCLIP · Multi-Modal-CelebA-HQ test (5,000) | ID similarity: 69.88 -> 73.24 | Improved |
| add entropy loss to the baseline (ablation) | ManiCLIP · Multi-Modal-CelebA-HQ test (5,000) | attribute accuracy: 83.60 -> 84.73 | Improved |
| add decoupling training to the baseline (ablation) | ManiCLIP · Multi-Modal-CelebA-HQ test (5,000) | attribute accuracy: 83.60 -> 84.38 | Improved |
| use group sampling instead of random sampling at Na=3 (ablation) | ManiCLIP · Multi-Modal-CelebA-HQ test (5,000) | FID: 35.94 -> 31.57 | Improved |
| use group sampling instead of random sampling at Na=1 (ablation) | ManiCLIP · Multi-Modal-CelebA-HQ test (5,000) | FID: 35.50 -> 34.01 | Improved |
| use group sampling instead of random sampling at Na=5 (ablation) | ManiCLIP · Multi-Modal-CelebA-HQ test (5,000) | FID: 37.80 -> 36.26 | Improved |
| increase group-sampled attribute number from 3 to 5 (ablation) | ManiCLIP · Multi-Modal-CelebA-HQ test (5,000) | FID: 31.57 -> 36.26 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TediGAN — TediGAN: Text-Guided Diverse Face Image Generation and Manipulation (CVPR 2021) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- PreciseControl: Enhancing Text-To-Image Diffusion Models with Fine-Grained Attribute Control (ECCV 2024 · similarity 0.89)
- SDGAN: Disentangling Semantic Manipulation for Facial Attribute Editing (AAAI 2024 · similarity 0.87)
- Text-conditional Attribute Alignment across Latent Spaces for 3D Controllable Face Image Synthesis (CVPR 2024 · similarity 0.86)
- CA-Edit: Causality-Aware Condition Adapter for High-Fidelity Local Facial Attribute Editing (AAAI 2025 · similarity 0.86)
- DreamSalon: A Staged Diffusion Framework for Preserving Identity-Context in Editable Face Generation (CVPR 2024 · similarity 0.86)
- DreamIdentity: Enhanced Editability for Efficient Face-Identity Preserved Image Generation (AAAI 2024 · similarity 0.84)
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.
