Language Segment-Anything is an open-source project that combines the power of instance segmentation and text prompts to generate masks for specific objects in images. Built on the recently released Meta model, Segment Anything Model 2, and the GroundingDINO detection model, it's an easy-to-use and effective tool for object detection and image segmentation.
Before installing lang-sam, please install PyTorch using the following command:
pip install torch==2.4.1 torchvision==0.19.1 --extra-index-url https://download.pytorch.org/whl/cu124
pip install -U git+https://github.com/luca-medeiros/lang-segment-anything.git
Or Clone the repository and install the required packages:
git clone https://github.com/luca-medeiros/lang-segment-anything && cd lang-segment-anything pip install -e .
Build and run the image.
git clone https://github.com/luca-medeiros/lang-segment-anything && cd lang-segment-anything docker build --tag lang-segment-anything:latest . docker run --gpus all -p 8000:8000 lang-segment-anything:latest
To run the gradio APP:
python app.py And open http://0.0.0.0:8000/gradio
Use as a library:
from PIL import Image
from lang_sam import LangSAM
model = LangSAM()
image_pil = Image.open("./assets/car.jpeg").convert("RGB")
text_prompt = "wheel."
results = model.predict([image_pil], [text_prompt])
If desired, arguments below can be passed for offline operations:
LangSAM(sam_ckpt_path, # path for segment anything model
gdino_model_ckpt_path, # path for grounding dino's model checkpoint
gdino_processor_ckpt_path # path for grounding dino's processor checkpoint
)
This project is based on/used the following repositories:
This project is licensed under the Apache 2.0 License
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4
