This repo is a usefull backend for an innovative project we've worked on in our journey through HackUPC 2024. By using lots of Inditex's data, we've created a new way to try on clothes based on a recommendation system that works with embeddings.
# Clone the repo
git clone https://github.com/HACKUPC-Try-on/backend-core
# Make sure to have poetry and Python 3.10 running and install the dependencies
poetry install --no-root
# Install Grounding DINO
mkdir -p packages && cd packages
git clone https://github.com/IDEA-Research/GroundingDINO.git
cd GroundingDINO
poetry run python setup.py install
# Weights
cd ../../
mkdir -p weights/dino && cd weights/dino
cp ../../packages/GroundingDINO/groundingdino/config/GroundingDINO_SwinT_OGC.py .
wget https://github.com/IDEA-Research/GroundingDINO/releases/download/v0.1.0-alpha/groundingdino_swint_ogc.pth
cd ..
mkdir -p weights/sam && cd weights/sam
wget https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth
# Run the API!
poetry run python uvicorn main:app
That's it! Our API is running on localhost, on port 8000!
If you want to contribute to our project, feel free to do!
Make sure to check out our CONTRIBUTING.md
file