Simple API for image color analysis.
This is a first attempt at using Rust, and largely a learning experience. Please feel free to contribute and teach me a few things.
Feel free to test out the project here
- Submit a hosted URL or file upload at upload page
- Allows for image url submission at predict endpoint
- Allows for form data submission at submit endpoint
- Parses base hex colors from server/src/colors.json
- Output ordered list of colors based on individual pixel analysis
- Runs on GAE Flexible custom runtime
- UI Option to predict based on middle of image (middle 50% of pixels)
make dev
make test
docker tag color_analyzer:latest us.gcr.io/rust-color-analyzer/color_analyzer
gcloud auth configure-docker
docker push us.gcr.io/rust-color-analyzer/color_analyzer
gcloud app deploy --image-url=us.gcr.io/rust-color-analyzer/color_analyzer app.yaml
- Add K-NN approach for determining dominant color
- Run all prediction operations asynchronously
- Explore Luma color pixels as an options