Research map / Image generation / Image restoration
Distillation-Guided Image Inpainting
Maitreya Suin, Kuldeep Purohit, A. N. Rajagopalan
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 cross-distillation (ablation) | Baseline · Places2 30-40% | PSNR: 22.36 -> 22.84 | Improved |
| add self-distillation (ablation) | Baseline · Places2 30-40% | PSNR: 22.36 -> 22.79 | Improved |
| combine self and cross distillation (ablation) | Baseline · Places2 30-40% | PSNR: 22.98 | Improved |
| add completion blocks to distilled encoder (ablation) | Net3 · Places2 30-40% | PSNR: 22.98 -> 23.54 | Improved |
| add completion blocks without distillation (ablation) | Baseline · Places2 30-40% | PSNR: 22.36 -> 22.54 | Improved |
| add attention transfer distillation (ablation) | Net6 · Places2 30-40% | PSNR: 23.62 -> 24.08 | Improved |
| add pixel-adaptive global-local fusion (ablation) | Net7 · Places2 30-40% | PSNR: 24.08 -> 24.21 | Improved |
| apply distillation to GC (ablation) | GC · Places2 30-40% | PSNR: 22.04 -> 22.74 | Improved |
| apply distillation to EC (ablation) | EC · Places2 30-40% | PSNR: 22.18 -> 22.62 | Improved |
| apply distillation to SF (ablation) | SF · Places2 30-40% | PSNR: 23.31 -> 23.70 | Improved |
| add attention transfer to contextual attention (ablation) | GC · Places2 30-40% | PSNR: 22.04 -> 22.41 | Improved |
| remove edge stage of EC and use distillation instead (ablation) | EC · Places2 30-40% | PSNR: 22.18 -> 22.06 | Mixed |
Closest papers by meaning
- Generative Image Inpainting With Contextual Attention (CVPR 2018 · similarity 0.89)
- Contextual-based Image Inpainting: Infer, Match, and Translate (ECCV 2018 · similarity 0.89)
- Deep Learning-Based Image and Video Inpainting: A Survey (IJCV 2024 · similarity 0.88)
- Deep Video Inpainting (CVPR 2019 · similarity 0.88)
- StructureFlow: Image Inpainting via Structure-Aware Appearance Flow (ICCV 2019 · similarity 0.88)
- Learning Prior Feature and Attention Enhanced Image Inpainting (ECCV 2022 · similarity 0.88)
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.
