metadata
license: apache-2.0
tags:
- segmentation
- prompted-segmentation
- clipseg
- drywall
drywall-clipseg
CLIPSeg fine-tune (HF CIDAS/clipseg-rd64-refined backbone, decoder + FiLM
unfrozen) for prompted binary segmentation on drywall imagery. Single
checkpoint covers two classes selected by text prompt:
"segment crack"→ wall-crack mask"segment taping area"→ drywall taping-seam mask
Test metrics (focal_dice loss, threshold 0.6)
| Task | Dice | mIoU | Precision | Recall |
|---|---|---|---|---|
| Crack | 0.672 | 0.531 | — | — |
| Taping | 0.727 | 0.587 | — | — |
Load + predict
from huggingface_hub import hf_hub_download
import torch
from src.models.clipseg_wrapper import CLIPSegFT
ckpt_path = hf_hub_download(repo_id="ravindrakapse/drywall-clipseg", filename="best.pt")
model = CLIPSegFT(pretrained="CIDAS/clipseg-rd64-refined").cuda()
state = torch.load(ckpt_path, map_location="cuda")
model.load_state_dict(state["model"])
model.eval()
See load_models.py for the full inference pipeline (letterbox + TTA + un-letterbox).