Research map / Image generation / Autoregressive and latent generation
Laplacian Patch-Based Image Synthesis
Joo Ho Lee, Inchang Choi, Min H. Kim
Check 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 Laplacian patch-based synthesis instead of Darabi et al.'s gradient-based method | Image completion · Figure 4 images (5) | runtime (seconds/image): 408.83 -> 111.62 | Improved |
| Use Laplacian patch-based synthesis instead of Wexler et al. | Image completion · Figure 4 images (5) | runtime (seconds/image): 107.33 -> 111.62 | Hurt |
| Replace Poisson reconstruction with DoG-based Laplacian pyramid reconstruction | Laplacian patch-based synthesis · Figure 4 images (5) | reconstruction runtime (seconds/image): 34.26 -> 0.02 | Improved |
| Add a nonzero Laplacian term to the correspondence distance (ablation) | Laplacian patch-based synthesis · Image completion example | high-frequency detail preservation | Improved |
| Set the Laplacian-term weight β to zero (ablation) | Laplacian patch-based synthesis · Image completion example | high-frequency detail preservation | Hurt |
| Use an upsampled Gaussian pyramid instead of the naïve Gaussian pyramid (ablation) | Patch-based image synthesis · Bungee example | low-level edge structure preservation | Improved |
| Use an upsampled Gaussian rather than the Gaussian pyramid for base-structure search (ablation) | Patch-based image synthesis · Image completion examples in Figure 9 | global structural coherence | Improved |
| Increase patch size across the tested 5-by-5, 7-by-7, and 9-by-9 settings (ablation) | Laplacian patch-based synthesis · Bungee example | low-level edge structure preservation | No clear effect |
Closest papers by meaning
- Lookup Table meets Local Laplacian Filter: Pyramid Reconstruction Network for Tone Mapping (NeurIPS 2023 · similarity 0.86)
- Laplacian Coordinates: Theory and Methods for Seeded Image Segmentation (TPAMI 2020 · similarity 0.85)
- Patched Denoising Diffusion Models For High-Resolution Image Synthesis (ICLR 2024 · similarity 0.85)
- PatchScaler: An Efficient Patch-Independent Diffusion Model for Image Super-Resolution (ICCV 2025 · similarity 0.84)
- LaMAR: Laplacian Pyramid for Multimodal Adaptive Super Resolution (Student Abstract) (AAAI 2024 · similarity 0.83)
- Exploring the Design Space of Diffusion Bridge Models (NeurIPS 2025 · similarity 0.82)
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.
