Research map / Image generation / Image restoration
MISF: Multi-Level Interactive Siamese Filtering for High-Fidelity Image Inpainting
Xiaoguang Li, Qing Guo, Di Lin, Ping Li, Wei Feng, Song Wang
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 |
|---|---|---|---|
| combine image-level and semantic filtering with interactive, input-conditioned kernels (ablation) | MISF · Places2 40%-60% mask | PSNR: 19.944 -> 20.044 | Improved |
| combine image-level and semantic filtering with interactive, input-conditioned kernels (ablation) | MISF · Places2 40%-60% mask | LPIPS: 0.2527 -> 0.2499 | Improved |
| use image-level filtering only instead of semantic filtering (ablation) | MISF · Places2 40%-60% mask | PSNR: 20.044 -> 13.773 | Hurt |
| remove dynamic filtering operations (ablation) | MISF · Places2 40%-60% mask | PSNR: 19.871 -> 20.044 | Hurt |
| add image-level predictive filtering to encoder-decoder output (ablation) | MISF · Places2 40%-60% mask | PSNR: 19.871 -> 19.898 | Improved |
| perform semantic filtering at F3 rather than F1 (ablation) | Sem-Filter · CelebA 40%-60% mask | PSNR: 21.360 -> 21.486 | Improved |
| perform semantic filtering at F3 rather than F2 (ablation) | Sem-Filter · CelebA 40%-60% mask | PSNR: 21.353 -> 21.486 | Improved |
| use MISF rather than JPGNet | image inpainting · Places2 40%-60% mask | PSNR: 19.884 -> 20.044 | Improved |
| use MISF rather than JPGNet | image inpainting · Dunhuang default | LPIPS: 0.0469 -> 0.0330 | Improved |
Later papers that compared with this method
3 later accepted papers report MISF as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- High-Fidelity Image Inpainting with Multimodal Guided GAN Inversion (IJCV 2025 · better on 2 of 2 shared results)
- SAIR: Learning Semantic-aware Implicit Representation (ECCV 2024 · better on 6 of 6 shared results)
- Cross-Image Context for Single Image Inpainting (NeurIPS 2022 · better on 8 of 8 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
- Semantic Image Inpainting With Deep Generative Models (CVPR 2017 · similarity 0.89)
- High-Resolution Image Inpainting with Iterative Confidence Feedback and Guided Upsampling (ECCV 2020 · similarity 0.88)
- Image Inpainting via Generative Multi-column Convolutional Neural Networks (NeurIPS 2018 · similarity 0.88)
- Towards Enhanced Image Inpainting: Mitigating Unwanted Object Insertion and Preserving Color Consistency (CVPR 2025 · similarity 0.87)
- DesignEdit: Unify Spatial-Aware Image Editing via Training-free Inpainting with a Multi-Layered Latent Diffusion Framework (AAAI 2025 · similarity 0.87)
- Paint by Inpaint: Learning to Add Image Objects by Removing Them First (CVPR 2025 · 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.
