Update README.md
Browse files
README.md
CHANGED
|
@@ -9,15 +9,22 @@ tags:
|
|
| 9 |
- fastsam
|
| 10 |
pipeline_tag: image-segmentation
|
| 11 |
---
|
|
|
|
| 12 |
# FastSAM ONNX Weights
|
|
|
|
| 13 |
This repository contains ONNX-optimized weights for **FastSAM (Fast Segment Anything Model)**, a real-time CNN-based instance segmentation model built upon the YOLOv8-Seg architecture.
|
| 14 |
By converting the PyTorch weights to standard ONNX format, these models enable real-time object proposal generation and mask decoding on both CPU and GPU edge devices without requiring the PyTorch runtime.
|
|
|
|
| 15 |
## Available Files
|
|
|
|
| 16 |
* **`FastSAM-s.onnx`**: Small YOLOv8-Seg backbone variant (`"s"`), optimized for high frame rate and edge deployment.
|
| 17 |
* **`FastSAM-x.onnx`**: Extra-large YOLOv8-Seg backbone variant (`"x"`), higher capacity proposal generation model.
|
|
|
|
| 18 |
---
|
| 19 |
## How to Use
|
|
|
|
| 20 |
The easiest way to load and run these models is through the **[spatialhub](https://github.com/spatialhub-ai/spatialhub)** Python library:
|
|
|
|
| 21 |
```python
|
| 22 |
from spatialhub import FastSAM
|
| 23 |
# Initialize FastSAM segmentor
|
|
@@ -29,7 +36,9 @@ result.visualize_mask("fastsam_output.png")
|
|
| 29 |
```
|
| 30 |
|
| 31 |
## Original Citation
|
|
|
|
| 32 |
If you use these models in academic work, please cite the original authors:
|
|
|
|
| 33 |
```bibtex
|
| 34 |
@article{zhao2023fast,
|
| 35 |
title={Fast Segment Anything},
|
|
|
|
| 9 |
- fastsam
|
| 10 |
pipeline_tag: image-segmentation
|
| 11 |
---
|
| 12 |
+
|
| 13 |
# FastSAM ONNX Weights
|
| 14 |
+
|
| 15 |
This repository contains ONNX-optimized weights for **FastSAM (Fast Segment Anything Model)**, a real-time CNN-based instance segmentation model built upon the YOLOv8-Seg architecture.
|
| 16 |
By converting the PyTorch weights to standard ONNX format, these models enable real-time object proposal generation and mask decoding on both CPU and GPU edge devices without requiring the PyTorch runtime.
|
| 17 |
+
|
| 18 |
## Available Files
|
| 19 |
+
|
| 20 |
* **`FastSAM-s.onnx`**: Small YOLOv8-Seg backbone variant (`"s"`), optimized for high frame rate and edge deployment.
|
| 21 |
* **`FastSAM-x.onnx`**: Extra-large YOLOv8-Seg backbone variant (`"x"`), higher capacity proposal generation model.
|
| 22 |
+
|
| 23 |
---
|
| 24 |
## How to Use
|
| 25 |
+
|
| 26 |
The easiest way to load and run these models is through the **[spatialhub](https://github.com/spatialhub-ai/spatialhub)** Python library:
|
| 27 |
+
|
| 28 |
```python
|
| 29 |
from spatialhub import FastSAM
|
| 30 |
# Initialize FastSAM segmentor
|
|
|
|
| 36 |
```
|
| 37 |
|
| 38 |
## Original Citation
|
| 39 |
+
|
| 40 |
If you use these models in academic work, please cite the original authors:
|
| 41 |
+
|
| 42 |
```bibtex
|
| 43 |
@article{zhao2023fast,
|
| 44 |
title={Fast Segment Anything},
|