Food Recognition Benchmark Starter Kit

screenshot of Food Recognition Benchmark Starter Kit

Food Recognition Benchmark Starter Kit

This repository is the main Food Recognition Benchmark template and Starter kit. Clone the repository to compete now!

Overview

The Food-Challenge Food Recognition Benchmark Starter Kit is a repository that provides a template and starter code for the Food Recognition Benchmark. The goal of this benchmark is to train models that can look at images of food items and detect the individual food items present in them. This ongoing benchmark consists of multiple rounds, each with its own tasks, datasets, and prizes. Participants can choose to participate in multiple rounds or single rounds. The repository contains the following:

  • mmdetection and detectron2 baselines for tackling the benchmark
  • Documentation on how to submit models to the leaderboard
  • Best practices and information on how models are evaluated
  • Starter code to help participants get started
  • Option to make submissions from Google Colab for resource-constrained users

Features

  • Baselines for mmdetection and detectron2
  • Documentation on submission process
  • Best practices and evaluation information
  • Starter code for participants to start working on the benchmark
  • Option to make submissions from Google Colab

Installation

To install the Food-Challenge Food Recognition Benchmark Starter Kit, follow the steps below:

  1. Clone the repository to your local machine:
git clone [repository-url]
  1. Install the required packages using pip:
pip install -r requirements.txt
  1. Install the required packages using apt (for necessary packages):
apt install -r apt.txt
  1. Run the code locally using the provided predict scripts:
  • For Detectron2 submission:
python predict_detectron2.py
  • For MMdetection submission:
python predict_mmdetection.py

Summary

The Food-Challenge Food Recognition Benchmark Starter Kit is a repository that provides the necessary tools and code to participate in the Food Recognition Benchmark. It includes baselines for popular object detection libraries, documentation on submission process and evaluation, best practices, and starter code. Participants can choose to make submissions from their local machines or from Google Colab. Overall, the starter kit aims to facilitate the participation process and encourage the development of food recognition models.