# Google Landmark Retrieval 2021 Competition Third Place Solution

arXiv: https://arxiv.org/abs/2110.04619 · HTML: https://mlsanity.com/paper/2110.04619/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 17 references that are in the map.
- First posted on arXiv 9 Oct 2021.

## Builds on (most cited first, 12 of 12)

- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [EfficientNetV2: Smaller Models and Faster Training](https://mlsanity.com/paper/2104.00298/) (2021)
- [Google Landmarks Dataset v2 -- A Large-Scale Benchmark for Instance-Level Recognition and Retrieval](https://mlsanity.com/paper/2004.01804/) (2020)
- [Large-scale Landmark Retrieval/Recognition under a Noisy and Diverse Dataset](https://mlsanity.com/paper/1906.04087/) (2019)
- [Supporting large-scale image recognition with out-of-domain samples](https://mlsanity.com/paper/2010.01650/) (2020)
- [Google Landmark Recognition 2020 Competition Third Place Solution](https://mlsanity.com/paper/2010.05350/) (2020)
- [Towards A Fairer Landmark Recognition Dataset](https://mlsanity.com/paper/2108.08874/) (2021)
- [1st Place Solution to Google Landmark Retrieval 2020](https://mlsanity.com/paper/2009.05132/) (2020)
- [Efficient large-scale image retrieval with deep feature orthogonality and Hybrid-Swin-Transformers](https://mlsanity.com/paper/2110.03786/) (2021)
- [Team JL Solution to Google Landmark Recognition 2019](https://mlsanity.com/paper/1906.11874/) (2019)
- [2nd Place Solution to Google Landmark Recognition Competition 2021](https://mlsanity.com/paper/2110.02638/) (2021)

## Abstract (arXiv)

We present our solutions to the Google Landmark Challenges 2021, for both the retrieval and the recognition tracks. Both solutions are ensembles of transformers and ConvNet models based on Sub-center ArcFace with dynamic margins. Since the two tracks share the same training data, we used the same pipeline and training approach, but with different model selections for the ensemble and different post-processing. The key improvement over last year is newer state-of-the-art vision architectures, especially transformers which significantly outperform ConvNets for the retrieval task. We finished third and fourth places for the retrieval and recognition tracks respectively.