Oktoberfest Food Dataset

TL;DR

This paper introduces the Oktoberfest Food Dataset, a comprehensive collection of annotated images and unlabeled footage from a German beer tent, aimed at advancing object detection for food and drink items.

Contribution

It provides a new, challenging dataset with annotations and trained models for object detection, specifically tailored for food and drink items in a festive setting.

Findings

Over 2,500 annotated objects across 15 categories

600GB of unlabeled video footage available

Trained models provided as benchmarks

Abstract

We release a realistic, diverse, and challenging dataset for object detection on images. The data was recorded at a beer tent in Germany and consists of 15 different categories of food and drink items. We created more than 2,500 object annotations by hand for 1,110 images captured by a video camera above the checkout. We further make available the remaining 600GB of (unlabeled) data containing days of footage. Additionally, we provide our trained models as a benchmark. Possible applications include automated checkout systems which could significantly speed up the process.