Research map / Image generation / Text-to-image diffusion
Improving Text-guided Object Inpainting with Semantic Pre-inpainting
Yifu Chen, Jingwen Chen, Yingwei Pan, Yehao Li, Ting Yao, Zhineng Chen, Tao Mei
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 reference adapter layer fed with masked-image CLIP features (ablation) | Finetuned Stable Inpainting · OpenImages-V6 (segmentation mask) | FID / Local FID: FID 5.15 -> 4.24; Local FID 8.80 -> 7.33 | Improved |
| add semantic inpainter outputs on top of reference adapter (ablation) | CAT-Diffusion · OpenImages-V6 (segmentation mask) | FID / Local FID: FID 4.24 -> 3.99; Local FID 7.33 -> 6.32 | Improved |
| add semantic inpainter and adapter on top of fine-tuned inpainting model (ablation) | Stable Inpainting · OpenImages-V6 (segmentation mask) | CLIP score: 0.262 -> 0.263 | No clear effect |
| semantic inpainter refines CLIP features of masked region | CAT-Diffusion · 10K images | cosine similarity to ground truth: 0.47 -> 0.65 | Improved |
| CAT-Diffusion versus SmartBrush | text-guided object inpainting · OpenImages-V6 (segmentation mask) | Local FID: 7.82 -> 6.32 | Improved |
| CAT-Diffusion versus SmartBrush on CLIP score | text-guided object inpainting · OpenImages-V6 (bounding box mask) | CLIP score: 0.266 -> 0.265 | No clear effect |
| CAT-Diffusion versus Stable Inpainting and SmartBrush zero-shot on MSCOCO | text-guided object inpainting · MSCOCO (bounding box mask) | Local FID: relative improvements of 42.1% and 20.7% | Improved |
| CAT-Diffusion versus SD-Inpaint in user study | text-guided object inpainting · OpenImages-V6 user study | Visual coherence (0-5): 2.6 -> 3.6 | Improved |
| add semantic inpainter and adapter | Stable Inpainting · inference time per image | seconds: 1.60 -> 1.84 | Hurt |
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.
- SmartBrush — SmartBrush: Text and Shape Guided Object Inpainting With Diffusion Model (CVPR 2023) · better on 8 of 9 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
- HD-Painter: High-Resolution and Prompt-Faithful Text-Guided Image Inpainting with Diffusion Models (ICLR 2025 · similarity 0.89)
- One Stone with Two Birds: A Null-Text-Null Frequency-Aware Diffusion Models for Text-Guided Image Inpainting (NeurIPS 2025 · similarity 0.89)
- Paint by Inpaint: Learning to Add Image Objects by Removing Them First (CVPR 2025 · similarity 0.89)
- Brush2Prompt: Contextual Prompt Generator for Object Inpainting (CVPR 2024 · similarity 0.89)
- MTADiffusion: Mask Text Alignment Diffusion Model for Object Inpainting (CVPR 2025 · similarity 0.89)
- Multi-Region Text-Driven Manipulation of Diffusion Imagery (AAAI 2024 · similarity 0.89)
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.
