The 2021 Image Similarity Dataset and Challenge [article]

Matthijs Douze and Giorgos Tolias and Ed Pizzi and Zoë Papakipos and Lowik Chanussot and Filip Radenovic and Tomas Jenicek and Maxim Maximov and Laura Leal-Taixé and Ismail Elezi and Ondřej Chum and Cristian Canton Ferrer
2022 arXiv   pre-print
This paper introduces a new benchmark for large-scale image similarity detection. This benchmark is used for the Image Similarity Challenge at NeurIPS'21 (ISC2021). The goal is to determine whether a query image is a modified copy of any image in a reference corpus of size 1~million. The benchmark features a variety of image transformations such as automated transformations, hand-crafted image edits and machine-learning based manipulations. This mimics real-life cases appearing in social media,
more » ... for example for integrity-related problems dealing with misinformation and objectionable content. The strength of the image manipulations, and therefore the difficulty of the benchmark, is calibrated according to the performance of a set of baseline approaches. Both the query and reference set contain a majority of "distractor" images that do not match, which corresponds to a real-life needle-in-haystack setting, and the evaluation metric reflects that. We expect the DISC21 benchmark to promote image copy detection as an important and challenging computer vision task and refresh the state of the art. Code and data are available at
arXiv:2106.09672v4 fatcat:cqeu7wkmp5djtac453t5ao5ix4