Research map / 3D vision / Face and shape
Human-Inspired Facial Sketch Synthesis with Dynamic Adaptation
Fei Gao, Yifan Zhu, Chang Jiang, Nannan Wang
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 depth map as auxiliary input only (ablation) | Pix2Pix base · FS2K test | FID: 18.34 -> 20.1 | No clear effect |
| add DySPADE dynamic normalization (ablation) | Pix2Pix + depth · FS2K test | FID: 20.1 -> 18.30 | Improved |
| add InfoACON dynamic activation in decoder (ablation) | Pix2Pix + depth + DySPADE · FS2K test | FID: 18.30 -> 17.41 | Improved |
| add DOG (deformable convolutions at top 2 decoder layers) (ablation) | Pix2Pix + depth + DySPADE + InfoACON · FS2K test | FID: 17.41 -> 15.06 | Improved |
| add DOG (ablation) | Model-D · FS2K test | SCOOT: 0.493 -> 0.575 | Improved |
| use InfoACON instead of metaACON (ablation) | HIDA generator · FS2K test | FID: 19.05 -> 17.41 | Improved |
| use metaACON or InfoACON instead of ReLU/LeakyReLU (ablation) | HIDA generator · FS2K test | FID: ReLU 18.30; metaACON 19.05; InfoACON 17.41 | Mixed |
| apply DOG at bottom decoder layers or all layers instead of top 2 (ablation) | HIDA generator · FS2K test | style fidelity | Hurt |
| use HIDA instead of GENRE and SCA-GAN | facial sketch synthesis · FS2K test | FID: 20.67 -> 15.06 (GENRE); 39.63 -> 15.06 (SCA-GAN) | Improved |
| use HIDA instead of SCA-GAN on structural similarity | facial sketch synthesis · FS2K test | FSIM: 0.782 -> 0.551 | Hurt |
| replace SPADE with DySPADE and InfoACON and add depth in CoCosNet | CoCosNet · MetFace | FID: 34.14 -> 22.62 | Improved |
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.
- FSGAN — FSGAN: Subject Agnostic Face Swapping and Reenactment (ICCV 2019) · better on 2 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
- SketchEvo: Leveraging Drawing Dynamics for Enhanced Image Synthesis (ICLR 2026 · similarity 0.87)
- Text to Sketch Generation with Multi-Styles (NeurIPS 2025 · similarity 0.85)
- One-Shot Reference-based Structure-Aware Image to Sketch Synthesis (AAAI 2025 · similarity 0.85)
- SketchAgent: Language-Driven Sequential Sketch Generation (CVPR 2025 · similarity 0.85)
- Neural 3D Strokes: Creating Stylized 3D Scenes with Vectorized 3D Strokes (CVPR 2024 · similarity 0.84)
- HyperLoRA: Parameter-Efficient Adaptive Generation for Portrait Synthesis (CVPR 2025 · 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.
