|
--- |
|
license: apache-2.0 |
|
--- |
|
|
|
# BEN - Background Erase Network |
|
|
|
BEN is a deep learning model designed to automatically remove backgrounds from images, producing both a mask and a foreground image. |
|
|
|
|
|
## Quick Start Code |
|
```python |
|
from BEN import BEN_Base |
|
from PIL import Image |
|
import torch |
|
|
|
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') |
|
model = BEN_Base().to(device).eval() |
|
model.loadcheckpoints("./BEN/BEN_Base.pth") |
|
|
|
image = Image.open("./image2.jpg") |
|
mask, foreground = model.inference(image) |
|
mask.save("./mask.png") |
|
foreground.save("./foreground.png") |
|
|
|
|
|
# BEN SOA Benchmarks on Disk 5k Eval |
|
|
|
### BEN_Base + BEN_Refiner (commercial model please contact us for more information): |
|
- MAE: 0.0283 |
|
- DICE: 0.8976 |
|
- IOU: 0.8430 |
|
- BER: 0.0542 |
|
- ACC: 0.9725 |
|
|
|
### BEN_Base: |
|
- MAE: 0.0331 |
|
- DICE: 0.8743 |
|
- IOU: 0.8301 |
|
- BER: 0.0560 |
|
- ACC: 0.9700 |
|
|
|
### MVANet (old SOA): |
|
- MAE: 0.0353 |
|
- DICE: 0.8676 |
|
- IOU: 0.8104 |
|
- BER: 0.0639 |
|
- ACC: 0.9660 |
|
|
|
## Features |
|
- Background removal from images |
|
- Generates both binary mask and foreground image |
|
- CUDA support for GPU acceleration |
|
- Simple API for easy integration |
|
|
|
## Installation |
|
1. Clone Repo |
|
2. Install requirements.txt |
|
|
|
|
