DORI

Script for the DORI algorithm (make sure to adjust the image data directory before running it):

python DORI_v2.py

Installation

Clone the FastSAM 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 ftfy regex tqdm
pip install git+https://github.com/openai/CLIP.git

Install Segment Anything:

pip install git+https://github.com/facebookresearch/segment-anything.git

Install more packages:

pip install opencv-python pycocotools matplotlib onnxruntime onnx

Toy Examples

To see FastSAM run on some example images, make sure to adjust the folder location for the inputFolder and outputFolder on line 22 and 23 of run_FastSAM_toy.py, then run:

python3 run_FastSAM_toy.py

To see SAM run on some example images, make sure to adjust the folder location for the inputFolder and outputFolder on line 33 and 34 of run_SAM_toy.py, then run:

python3 run_SAM_toy.py

To see CLIP run on some example images, run the notebook:

CLIP_example.ipynb