image-inpaint-region
Local circular region inpainting with LaMa (IOPaint). Uses IOPaint default InpaintRequest and mask handling. Use for 局部é‡ç»?/ 局部修å¤?/ inpaint when circle center and radius cx,cy,r are known.
godot-fun · 已有 AI 审核记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add godot-fun/godot-agent --skill image-inpaint-region技能说明
name: image-inpaint-region description: >- Local circular region inpainting with LaMa (IOPaint). Uses IOPaint default InpaintRequest and mask handling. Use for 局部é‡ç»?/ 局部修å¤?/ inpaint when circle center and radius cx,cy,r are known.
Image Inpaint Region (LaMa)
Local circular region inpainting with IOPaint LaMa. Builds a circular mask from --region cx,cy,r, then runs the same inference path as iopaint run / the IOPaint API.
Both --image and --region cx,cy,r are required.
Rules
When this skill applies, read and follow skill-dependency-manager �run scripts as documented, install missing tools into .dependency/.
- Run
inpaint.pythrough theiopaintmanifest entry (.dependency/iopaint/.venv/). Never use hostpython,py,python3, or any interpreter outside.dependency/. - Do not hand-write equivalent inpaint commands �use the bundled script.
- Single file only. Pass one image with
--image; directories are not supported. - Require region first. If
--region cx,cy,ris missing, inspect the image (or ask the user) before running. - Use IOPaint defaults for inference config unless the user explicitly asks otherwise �do not invent custom mask blur, compositing, or hd_strategy overrides in the script.
- Pass the input path as-is. Output goes to
<image-dir>/image-inpaint-region/by default �no path rewriting. - Never overwrite source files. Output lands in
image-inpaint-region/or--output.
Setup (first run)
IOPaint needs Python 3.11 and PyTorch. From project root:
.dependency/python-3.11/python.exe -m venv .dependency/iopaint/.venv
.dependency/iopaint/.venv/Scripts/python.exe -m pip install torch torchvision --index-url https://download.pytorch.org/whl/cu118
.dependency/iopaint/.venv/Scripts/python.exe -m pip install iopaint
CPU-only: skip the CUDA torch line and pip install iopaint (pulls CPU torch). Use bin/python on Unix.
Register in .dependency/manifest.json:
"iopaint": {
"populated": true,
"bin": ".dependency/iopaint/.venv/Scripts/python.exe"
}
LaMa weights (big-lama.pt) download on first run.
Quick Start
Both --image and --region are required:
# image/foo.png �image/image-inpaint-region/foo.png
# center (148, 248), radius 48
.dependency/iopaint/.venv/Scripts/python.exe .ai/image-inpaint-region/inpaint.py --image image/foo.png --region 148,248,48
GPU inference:
.dependency/iopaint/.venv/Scripts/python.exe .ai/image-inpaint-region/inpaint.py --image image/foo.png --region 148,248,48 --device cuda
Custom output path:
.dependency/iopaint/.venv/Scripts/python.exe .ai/image-inpaint-region/inpaint.py --image image/foo.png --region 148,248,48 -o image/foo_inpainted
How it works
- Open the image as RGB (same as
iopaintbatch processing). - Draw a filled circle mask from
--region cx,cy,r. - Binarize mask at 127 (same as
iopaint.batch_processing/ API). - Call
ModelManagerwithInpaintRequest()defaults. - Write PNG via
pil_to_bytes(..., "png", 100, ...).
IOPaint defaults used
These come from InpaintRequest() and iopaint run �the script does not override them:
| Setting | Default | Notes |
|---|---|---|
--model | lama | Same as iopaint run |
--device | cpu | Same as iopaint run / start; use --device cuda when GPU is available |
hd_strategy | Crop | LaMa erase-model preprocessing |
hd_strategy_crop_trigger_size | 800 | Crop when long side > 800 px |
hd_strategy_crop_margin | 128 | Margin around mask for crop strategy |
| Mask threshold | 127 | >= 127 �255, else 0 |
The skill only adds --region cx,cy,r so you do not need a separate mask file.
Options
| Option | Default | Notes |
|---|---|---|
--image | Required | Single supported image file |
--region | Required | cx,cy,r �center x, center y, radius in pixels |
| Output | <image-dir>/image-inpaint-region/<name>.png | Use -o / --output for custom file or directory |
--model | lama | IOPaint model name |
--device | cpu | cpu or cuda (falls back via IOPaint check_device) |
Supported inputs: .png, .jpg, .jpeg, .webp, .gif, .bmp, .tif, .tiff, .avif, .ico.
Agent Workflow
- Confirm source �user path only; do not copy into
image/or use chat attachment cache. - Confirm region �inspect the image and determine circle center
(cx, cy)and radiusrbefore running. - One file per run �process one image, verify the result, then repeat for additional files if needed.
- Inspect �check the inpainted area; leftover content �larger
r; nearby art eaten �smallerror move center. - Revert �delete the output file or
git restore; sources are never modified.
Agent Notes
- Use the bundled script, not hand-written IOPaint CLI with separate mask files unless the user already has a mask image.
- Missing iopaint venv �populate
.dependency/per skill-dependency-manager, retry same command. - Do not copy, move, or replace the source with inpainted output �tell the user where the output file is.
- For full IOPaint UI / brush mask editing, use the upstream
iopaint startworkflow instead.
Troubleshooting
| Issue | Fix |
|---|---|
iopaint missing / not populated | Follow Setup; update manifest |
Missing --region | Inspect image; both --image and --region cx,cy,r are required |
Invalid --region | Must be three numbers: center x, center y, radius (> 0) |
Directory passed to --image | Run once per file; this skill accepts image files only |
| Output already exists | Delete the existing output or choose a different -o path |
| CUDA not available | IOPaint falls back to cpu via check_device |
| Target still vi |