Upload README.md

README.md

@@ -1,5 +1,14 @@
 ---
 title: Profsam Fire Demo
+license: agpl-3.0
+library_name: ultralytics
+pipeline_tag: object-detection
+tags:
+- yolo
+- yolo11
+- fire-detection
+- computer-vision
+- realtime
 emoji: 👁
 colorFrom: blue
 colorTo: yellow
@@ -7,8 +16,117 @@ sdk: gradio
 sdk_version: 5.49.1
 app_file: app.py
 pinned: false
-license: agpl-3.0
 short_description: This hosts the ProFSAM's Yolov11n fire detector model
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# ProFSAM Fire Detector (YOLOv11n) — Gradio Demo
+
+Real-time **fire object detection** using the YOLOv11n checkpoint from the paper:
+
+> **Promptable Fire Segmentation: Unleashing SAM2's Potential for Real-Time Mobile Deployment with Strategic Bounding Box Guidance**  
+> Paper: https://arxiv.org/abs/2510.21782  
+> Code: https://github.com/UEmmanuel5/ProFSAM  
+> Model weights on the Hub: https://huggingface.co/UEmmanuel5/ProFSAM-Fire-Detector
+
+---
+
+## 👟 Try it
+- Upload an image (JPG/PNG).  
+- The app returns the same image with **bounding boxes** over detected fire regions.
+
+---
+
+## 🔧 How it works
+This Space loads the published checkpoint via the Hub and runs Ultralytics YOLOv11n inference:
+
+- Loads: `UEmmanuel5/ProFSAM-Fire-Detector` → `Fire_best.pt`
+- Inference size: `imgsz=640`, default confidence `0.3`
+- Outputs: annotated image with bounding boxes
+
+---
+
+## 📦 Files in this Space
+```
+
+app.py
+requirements.txt
+README.md
+
+```
+
+- **`requirements.txt`**
+```
+
+ultralytics
+gradio
+huggingface_hub
+
+````
+- **`app.py`** (summary)
+- Downloads `Fire_best.pt` from the Hub with `hf_hub_download`
+- Loads the checkpoint into `YOLO(...)`
+- Defines a `detect(img)` function and serves a Gradio UI
+
+---
+
+## 🖥️ Run locally (optional)
+1) Create a virtual env (recommended), then:
+```bash
+pip install -r requirements.txt
+````
+
+2. Run the app:
+
+```bash
+# Force CPU if your local GPU is not supported
+Then run:
+python app.py
+```
+
+3. Open the URL printed by Gradio (something like [http://127.0.0.1:xxxx](http://127.0.0.1:xxxx)) and upload an image.
+
+---
+
+## ⚖️ License
+
+* **Model weights** (`Fire_best.pt`): **AGPL-3.0** (Ultralytics-trained).
+* **Space code** (this repo): **Apache-2.0**.
+
+---
+
+## 🙏 Acknowledgements
+
+* Ultralytics YOLO11
+* SAM / SAM2 ecosystem and community
+
+---
+
+## 📚 Citation
+
+If this demo or the model is useful in your research, please cite:
+
+**Manuscript**
+
+```bibtex
+@inproceedings{profsam2025,
+  author    = {Emmanuel U. Ugwu and Xinming Zhang},
+  title     = {Promptable Fire Segmentation: Unleashing SAM2's Potential for Real-Time Mobile Deployment with Strategic Bounding Box Guidance},
+  booktitle = {ICIGP '26},
+  year      = {2026},
+  address   = {Wuhan, China},
+  month     = jan,
+  note      = {to appear}
+}
+```
+
+**Model/Code**
+
+```bibtex
+@software{profsam2025,
+  author    = {Ugwu, Emmanuel U. and Zhang, Xinming},
+  title     = {Promptable Fire Segmentation: Unleashing SAM2’s Potential for Real-Time Mobile Deployment with Strategic Bounding Box Guidance},
+  year      = {2025},
+  doi       = {10.5281/zenodo.17340313},
+  url       = {https://doi.org/10.5281/zenodo.17340313}
+}
+```