Research map / Image generation / Editing and style transfer
Disentangling Structure and Aesthetics for Style-Aware Image Completion
Andrew Gilbert, John Collomosse, Hailin Jin, Brian Price
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 a learned structure unary term to the no-style baseline (ablation) | Style-aware image completion · BAM! | SSIM: 0.85 -> 0.86 | Improved |
| Add the style term to structure-aware patch selection (ablation) | Style-aware image completion · BAM! | SSIM: 0.86 -> 0.87 | Improved |
| Pre-stylize candidate patches before MRF optimization (ablation) | Structure- and style-aware patch selection · BAM! | SSIM: 0.87 -> 0.91 | Improved |
| Use adaptive stylization during composition instead of pre-stylization (ablation) | Structure- and style-aware patch selection · BAM! | SSIM: 0.91 -> 0.94 | Improved |
| Add the structure unary term to the baseline (ablation) | Style-aware image completion · Places2 examples from [10] | SSIM: 0.76 -> 0.79 | Hurt |
| Add the style term to structure-aware selection (ablation) | Style-aware image completion · Places2 examples from [10] | SSIM: 0.79 -> 0.81 | Improved |
| Use adaptive stylization rather than pre-stylization (ablation) | Structure- and style-aware patch selection · Places2 examples from [10] | SSIM: 0.85 -> 0.90 | Improved |
| Use the proposed method instead of PatchMatch | Image completion · BAM! | SSIM: 0.86 -> 0.94 | Improved |
Closest papers by meaning
- Training-and-Prompt-Free General Painterly Harmonization via Zero-Shot Disentenglement on Style and Content References (AAAI 2025 · similarity 0.87)
- Prune and Repaint: Content-Aware Image Retargeting for any Ratio (NeurIPS 2024 · similarity 0.86)
- Yuan: Yielding Unblemished Aesthetics Through a Unified Network for Visual Imperfections Removal in Generated Images (AAAI 2025 · similarity 0.86)
- Progressive Painterly Image Harmonization from Low-Level Styles to High-Level Styles (AAAI 2024 · similarity 0.86)
- Paint by Inpaint: Learning to Add Image Objects by Removing Them First (CVPR 2025 · similarity 0.86)
- Painterly Image Harmonization by Learning from Painterly Objects (AAAI 2024 · similarity 0.85)
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.
