Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
language: en
|
| 4 |
+
library_name: segmentation-models-pytorch
|
| 5 |
+
tags:
|
| 6 |
+
- segmentation
|
| 7 |
+
- pytorch
|
| 8 |
+
- computer-vision
|
| 9 |
+
- leaf-segmentation
|
| 10 |
+
- interactive-segmentation
|
| 11 |
+
datasets:
|
| 12 |
+
- LeafNet75/Annotated_Benchmarks400
|
| 13 |
+
---
|
| 14 |
+
# Tomato-leaf-Segmentation
|
| 15 |
+
This is an interactive leaf segmentation model created by **Subh775**.
|
| 16 |
+
|
| 17 |
+
## Model Description
|
| 18 |
+
This model is a U-Net with a lightweight MobileNetV2 backbone, designed for interactive segmentation. It takes a 4-channel input (RGB image + user scribble) and outputs a binary segmentation mask.
|
| 19 |
+
This model was trained on the `LeafNet75/Annotated_Benchmarks400` dataset.
|
| 20 |
+
|
| 21 |
+
## Training Performance
|
| 22 |
+
The model was trained for 60 epochs with early stopping (patience=10).
|
| 23 |
+
|
| 24 |
+
### Final Validation Metrics:
|
| 25 |
+
- **Loss:** 0.0261
|
| 26 |
+
- **Dice Score:** 0.9475
|
| 27 |
+
- **IoU (Jaccard):** 0.9475
|
| 28 |
+
- **Precision:** 0.9711
|
| 29 |
+
- **Recall:** 0.9750
|
| 30 |
+
|
| 31 |
+
### Training History
|
| 32 |
+
|
| 33 |
+
|
| 34 |
+
### Sample Predictions
|
| 35 |
+
|