Research map / Label-efficient and robust learning / Domain adaptation and generalization
One-Shot Generative Domain Adaptation in 3D GANs
Ziqiang Li, Yi Wu, Chaoyue Wang, Xue Rui, Bin Li
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 |
|---|---|---|---|
| use two-step progressive fine-tuning instead of Tri-D only (ablation) | 3D-Adapter · Cartoon | FID: 141.6 -> 132.6 | Improved |
| use two-step progressive fine-tuning instead of G2 only (ablation) | 3D-Adapter · Cartoon | FID: 149.1 -> 132.6 | Improved |
| use two-step progressive fine-tuning instead of joint G2 and Tri-D tuning (ablation) | 3D-Adapter · Cartoon | FID: 171.2 -> 132.6 | Improved |
| use two-step progressive fine-tuning instead of joint tuning (ablation) | 3D-Adapter · Sketches | FID: 215.3 -> 59.59 | Improved |
| add target distribution learning loss (ablation) | 3D-Adapter · Cartoon | FID: 213.9 -> 132.6 | Improved |
| add target distribution learning loss (ablation) | 3D-Adapter · Sketches | FID: 307.6 -> 59.6 | Improved |
| add domain direction loss (ablation) | 3D-Adapter · Cartoon | FID: 148.6 -> 132.6 | Improved |
| add image-level structure maintenance loss (ablation) | 3D-Adapter · Cartoon | FID: 139.2 -> 132.6 | Improved |
| add feature-level structure maintenance loss (ablation) | 3D-Adapter · Cartoon | FID: 138.1 -> 132.6 | Improved |
| 3D-Adapter versus DiFa | 3D-Adapter · Sketches | FID: 219.4 -> 59.59 | Improved |
| 3D-Adapter versus DiFa on image quality quality in user study | 3D-Adapter · user study | preference: 21.09% -> 78.91% | Improved |
| 3D-Adapter versus DiFa on style similarity | 3D-Adapter · user study | style similarity preference: 55.76% -> 44.24% | Hurt |
Closest papers by meaning
- One-Shot Generative Domain Adaptation (ICCV 2023 · similarity 0.90)
- Generalized One-shot Domain Adaptation of Generative Adversarial Networks (NeurIPS 2022 · similarity 0.90)
- Towards Diverse and Faithful One-shot Adaption of Generative Adversarial Networks (NeurIPS 2022 · similarity 0.89)
- Transferring GANs: generating images from limited data (ECCV 2018 · similarity 0.88)
- Mind the Gap: Domain Gap Control for Single Shot Domain Adaptation for Generative Adversarial Networks (ICLR 2022 · similarity 0.87)
- DiffusionGAN3D: Boosting Text-guided 3D Generation and Domain Adaptation by Combining 3D GANs and Diffusion Priors (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.
