EfficientSAM

XetHub hosted fork of EfficientSAM using the GitHub plugin.

EfficientSAM: Leveraged Masked Image Pretraining for Efficient Segment Anything

News

[Dec.5 2023] We release the torchscript version of EfficientSAM and share the Google colab notebook.

Online Demo & Examples

Online demo and examples can be found in the project page.

EfficientSAM Instance Segmentation Examples

<img width="848" alt="Screenshot 2023-12-06 at 1 16 13 PM" src="https://github.com/xetdata/EfficientSAM/assets/801507/6a4586e6-94cc-4eea-94e7-e78a6f57dcfe">

Using the models

The model files live in this repo itself:

• EfficientSAM-S:models/efficientsam_s_gpu.jit
• EfficientSAM-Ti: models/efficientsam_ti_gpu.jit

<img width="971" alt="Screenshot 2023-12-06 at 1 19 26 PM" src="https://github.com/xetdata/EfficientSAM/assets/801507/72374016-2b6f-4617-881a-9b7b45059a35">

Instructions

1. Install the git-xet extension.

2. Clone the repo:

# Using SSH
git clone git@github.com:xetdata/EfficientSAM.git

cd EfficientSAM/

2. Activate your Python virtual environment and install the requirements:

pip install -r requirements.txt

3. Start Jupyter Notebook:

# jupyter lab
jupyter lab

# jupyter notebook
jupyter notebook

4. You can directly use EfficientSAM:

import torch

efficientsam = torch.jit.load(models/efficientsam_s_gpu.jit)

Google Colab

You can also use this Google Colab notebook

Acknowledgement

• SAM
• MobileSAM
• FastSAM
• U-2-Net

If you're using EfficientSAM in your research or applications, please cite using this BibTeX:

@article{xiong2023efficientsam,
  title={EfficientSAM: Leveraged Masked Image Pretraining for Efficient Segment Anything},
  author={Yunyang Xiong, Bala Varadarajan, Lemeng Wu, Xiaoyu Xiang, Fanyi Xiao, Chenchen Zhu, Xiaoliang Dai, Dilin Wang, Fei Sun, Forrest Iandola, Raghuraman Krishnamoorthi, Vikas Chandra},
  journal={arXiv:2312.00863},
  year={2023}
}