Fast Segment Anything

[Paper] [Web Demo] [Colab demo] [Model Zoo] [BibTeX]

The Fast Segment Anything Model(FastSAM) is a CNN Segment Anything Model trained by only 2% of the SA-1B dataset published by SAM authors. The FastSAM achieve a comparable performance with the SAM method at 50× higher run-time speed.

Installation

Clone the repository locally:

git clone https://github.com/CASIA-IVA-Lab/FastSAM.git

Create the conda env. The code requires python>=3.7, as well as pytorch>=1.7 and torchvision>=0.8. Please follow the instructions here to install both PyTorch and TorchVision dependencies. Installing both PyTorch and TorchVision with CUDA support is strongly recommended.

conda create -n FastSAM python=3.9
conda activate FastSAM

Install the packages:

cd FastSAM
pip install -r requirements.txt

Install clip:

pip install git+https://github.com/openai/CLIP.git

Getting Started

First download a model checkpoint.

Then, you can run the scripts to try the everything mode and three prompt modes.

# Everything mode
python Inference.py --model_path ./weights/FastSAM.pt --img_path ./images/dogs.jpg

# text prompt
python Inference.py --model_path ./weights/FastSAM.pt --img_path ./images/dogs.jpg  --text_prompt "the yellow dog"

# box prompt
python Inference.py --model_path ./weights/FastSAM.pt --img_path ./images/dogs.jpg --box_prompt [570,200,230,400]

# points prompt
python Inference.py --model_path ./weights/FastSAM.pt --img_path ./images/dogs.jpg  --point_prompt "[[520,360],[620,300]]" --point_label "[1,0]"

You are also welcomed to try our Colab demo: FastSAM_example.ipynb.

Different Inference Options

We provide various options for different purposes, details are in MORE_USAGES.md.

Web demo

In the web demo, you can upload your own image, select input size from 512~1024, and choose whether to visualize in high quality. High quality visualization additionally shows more easily observable split edges. The web demo only supports Everything Mode now, other modes will try to support in the future.

Model Checkpoints

Two model versions of the model are available with different sizes. Click the links below to download the checkpoint for the corresponding model type.

• default or FastSAM: YOLOv8x based Segment Anything Model.
• FastSAM-s: YOLOv8s based Segment Anything Model.

Results

All result were tested on a single NVIDIA GeForce RTX 3090.

1. Inference time

Running Speed under Different Point Prompt Numbers(ms).

2. Memory usage

3. Zero-shot Transfer Experiments

Edge Detection

Test on the BSDB500 dataset.

Object Proposals

COCO

LVIS

bbox AR@1000

Instance Segmentation On COCO 2017

4. Performance Visulization

Several segmentation results:

Natural Images

Text to Mask

5.Downstream tasks

The results of several downstream tasks to show the effectiveness.

Anomaly Detection

Salient Object Detection

Building Extracting

License

The model is licensed under the Apache 2.0 license.

Acknowledgement

• Segment Anything provides the SA-1B dataset and the base codes.
• YOLOv8 provides codes and pre-trained models.
• YOLACT provides powerful instance segmentation method.
• Grounded-Segment-Anything provides a useful web demo template.

Citing FastSAM

If you find this project useful for your research, please consider citing the following BibTeX entry.

@misc{zhao2023fast,
      title={Fast Segment Anything}, 
      author={Xu Zhao and Wenchao Ding and Yongqi An and Yinglong Du and Tao Yu and Min Li and Ming Tang and Jinqiao Wang},
      year={2023},
      eprint={2306.12156},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}