Food Image Segmentation & Nutrition Analysis (Django)

A Django web app where users can register, upload food photos, and receive:

• food segmentation output image
• ingredient area estimates (JSON)
• nutrition analysis computed from a reference CSV

This repo is a portfolio snapshot of an older project; the original model inference ran on a separate machine.

Tech stack

• Django
• Paramiko (SSH/SFTP) for remote inference orchestration
• Remote inference via docker exec ... (configurable)

Security note

No credentials are stored in this repo.
If you want to use the original remote-inference flow, provide secrets via environment variables (see .env.example).

Quick start (demo mode)

Demo mode skips SSH and uses local placeholder outputs so the app can be started without access to the inference machine.

python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate
pip install -r requirements.txt

cp .env.example .env
# Edit .env (INFERENCE_MODE=demo is the default in the example)

python manage.py migrate
python manage.py runserver

Then open http://127.0.0.1:8000/

Remote inference

Set INFERENCE_MODE=remote and fill the INFERENCE_* variables in your .env.

At minimum:

• INFERENCE_SSH_HOST, INFERENCE_SSH_PORT, INFERENCE_SSH_USER, INFERENCE_SSH_PASSWORD
• INFERENCE_REMOTE_UPLOAD_PATH, INFERENCE_REMOTE_SEGMENTED_PATH, INFERENCE_REMOTE_JSON_PATH
• optionally INFERENCE_DOCKER_COMMAND