Food Recognition Benchmark - Starter Kit
This repository is the main Food Recognition Benchmark template and Starter kit. Clone the repository to compete now!
This repository contains:
mmdetection
,detectron2
andmatterport-maskrcnn
baselines for tackling this benchmark- Documentation on how to submit your models to the leaderboard
- The procedure for best practices and information on how we evaluate your agent, etc.
- Starter code for you to get started!
NOTE: If you are resource-constrained or would not like to setup everything in your system, you can make your submission from inside Google Colab too. Check out the beta version of the Notebook.
The goal of this benchmark is to train models which can look at images of food items and detect the individual food items present in them. This is an ongoing, multi-round benchmark. At each round, the specific tasks and / or datasets will be updated, and each round will have its own prizes. You can participate in multiple rounds, or in single rounds.
This data set has been annotated with respect to segmentation, classification (mapping the individual food items onto an ontology of Swiss Food items), and weight/volume estimation.
πͺ Getting Started
π₯ Participation
𧩠Repository Structure
π Submission
π Important Links
This repository contains prediction codebase for mmdetection
, detectron2
and random agents.
# Clone the repository
git clone https://github.com/AIcrowd/food-recognition-challenge-starter-kit
cd food-recognition-challenge-starter-kit
# Install dependencies
pip install -r requirements.txt
# Download the dataset, and place it in `data/images/`
# Run model locally
./run.sh
This will generate predictions.json
file in your data/
directory.
Before we do a deep dive into submissions. Check which user persona suits you the best!
Quick Participation π | Active Participation π¨βπ» |
---|---|
You need to upload prediction json files | You need to submit code (and AIcrowd evaluators runs the code to generate predictions) |
Scores are computed on 40% of the publicly released test set | Scores are computed on 100% of the publicly released test set + 40% of the (unreleased) extended test set |
You are not eligible for the final leaderboard (and prizes) | You are eligible for the final leaderboard and prizes |
The flow for active participation look as follows:
File | Description |
---|---|
aicrowd.json |
A configuration file used to identify the challenge and resources needed for evaluation |
apt.txt |
List of packages that should be installed (via apt ) for your code to run |
requirements.txt |
List of python packages that should be installed (via pip ) for your code to run |
predict.py |
Entry point to your model |
File | Description |
---|---|
score.py |
Helps your generate score for your run locally |
utils/ |
Directory containing some useful scripts and notebooks |
utils/requirements_detectron2.txt |
A sample requirements.txt file for using detectron2 |
utils/requirements_mmdetection.txt |
A sample requirements.txt file for using mmdetection |
As promised, we will keep it quick for you. Participating is as simple as:
- Generate your predictions using the starter kit
- Upload
predictions.json
on the benchmark website - Get scores, iterate, improve! πͺ
- Prepare your runtime environment
- Make submissions by pushing your code repository
- Get scores, more scores π, iterate faster, improve faster! πͺ
More details for active participation in present in SUBMISSION.md
- πͺ Β Challenge Page: https://www.aicrowd.com/challenges/food-recognition-benchmark-2022
- π£οΈ Β Discussion Forum: https://www.aicrowd.com/challenges/food-recognition-benchmark-2022/discussion
- π Β Leaderboard: https://www.aicrowd.com/challenges/food-recognition-benchmark-2022/leaderboards
- Resources - Round 1
- Resources - Round 2
- Resources - Round 3
- Participant contributions
- External resources: