Image Segmentation
Transformers
English
clipseg
segmentation
construction
drywall
quality-assurance
text-conditioned
binary-mask
Instructions to use youngPhilosopher/drywall-qa-clipseg with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use youngPhilosopher/drywall-qa-clipseg with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-segmentation", model="youngPhilosopher/drywall-qa-clipseg")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("youngPhilosopher/drywall-qa-clipseg", dtype="auto") - Notebooks
- Google Colab
- Kaggle
| """Dataset download instructions. | |
| Both datasets must be downloaded manually from Roboflow Universe in COCO format. | |
| The Roboflow API cannot be used because the cracks dataset (cracks-3ii36) has | |
| no generated versions β the owner never created an exportable version. | |
| Download locations: | |
| - Taping: https://universe.roboflow.com/objectdetect-pu6rn/drywall-join-detect | |
| β Export as COCO, place under data/raw/taping/ | |
| - Cracks: https://universe.roboflow.com/fyp-ny1jt/cracks-3ii36 | |
| β Export as COCO, place under data/raw/cracks/ | |
| Expected structure after download: | |
| data/raw/ | |
| βββ taping/ | |
| β βββ train/ | |
| β β βββ _annotations.coco.json | |
| β β βββ *.jpg | |
| β βββ valid/ | |
| β βββ _annotations.coco.json | |
| β βββ *.jpg | |
| βββ cracks/ | |
| βββ train/ | |
| βββ _annotations.coco.json | |
| βββ *.jpg | |
| """ | |