ODOR: The ICPR2022 ODeuropa Challenge on Olfactory Object Recognition

TL;DR

The Odeuropa Challenge on Olfactory Object Recognition aims to advance object detection in historical artworks by encouraging innovative methods like domain adaptation and few-shot learning, supported by a specialized annotated dataset.

Contribution

It introduces a new dataset and challenge focused on detecting objects in artworks, addressing challenges like style variation and limited training data for certain objects.

Findings

Dataset of 2647 artworks with 20,120 annotations provided

Challenge promotes development of domain adaptation and few-shot learning methods

Benchmark results will evaluate approaches on a private test set

Abstract

The Odeuropa Challenge on Olfactory Object Recognition aims to foster the development of object detection in the visual arts and to promote an olfactory perspective on digital heritage. Object detection in historical artworks is particularly challenging due to varying styles and artistic periods. Moreover, the task is complicated due to the particularity and historical variance of predefined target objects, which exhibit a large intra-class variance, and the long tail distribution of the dataset labels, with some objects having only very few training examples. These challenges should encourage participants to create innovative approaches using domain adaptation or few-shot learning. We provide a dataset of 2647 artworks annotated with 20 120 tightly fit bounding boxes that are split into a training and validation set (public). A test set containing 1140 artworks and 15 480 annotations is kept private for the challenge evaluation.