UniUD Submission to the EPIC-Kitchens-100 Multi-Instance Retrieval Challenge 2023

TL;DR

This paper details a submission to the EPIC-Kitchens-100 Multi-Instance Retrieval Challenge 2023, involving model ensembling and achieving competitive retrieval scores on the leaderboard.

Contribution

The paper introduces a model ensembling approach trained with different loss functions for multi-instance retrieval in egocentric videos.

Findings

Achieved 56.81% nDCG score

Achieved 42.63% mAP score

Utilized ensembling of models trained on limited data

Abstract

In this report, we present the technical details of our submission to the EPIC-Kitchens-100 Multi-Instance Retrieval Challenge 2023. To participate in the challenge, we ensembled two models trained with two different loss functions on 25% of the training data. Our submission, visible on the public leaderboard, obtains an average score of 56.81% nDCG and 42.63% mAP.