Google Landmark Recognition 2020 Competition Third Place Solution
2020-10-11Code Available1· sign in to hype
Qishen Ha, Bo Liu, Fuxu Liu, Peiyuan Liao
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/haqishen/Google-Landmark-Recognition-2020-3rd-Place-SolutionOfficialpytorch★ 147
- github.com/MrTornado24/CS498_DL_Projectpytorch★ 1
Abstract
We present our third place solution to the Google Landmark Recognition 2020 competition. It is an ensemble of global features only Sub-center ArcFace models. We introduce dynamic margins for ArcFace loss, a family of tune-able margin functions of class size, designed to deal with the extreme imbalance in GLDv2 dataset. Progressive finetuning and careful postprocessing are also key to the solution. Our two submissions scored 0.6344 and 0.6289 on private leaderboard, both ranking third place out of 736 teams.