# 6th Place Solution to Google Universal Image Embedding

arXiv: https://arxiv.org/abs/2210.09377 · HTML: https://mlsanity.com/paper/2210.09377/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 17 references that are in the map.
- First posted on arXiv 17 Oct 2022.

## Builds on (most cited first, 11 of 11)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [Robust fine-tuning of zero-shot models](https://mlsanity.com/paper/2109.01903/) (2021)
- [Deep Metric Learning via Lifted Structured Feature Embedding](https://mlsanity.com/paper/1511.06452/) (2015)
- [Google Landmarks Dataset v2 -- A Large-Scale Benchmark for Instance-Level Recognition and Retrieval](https://mlsanity.com/paper/2004.01804/) (2020)
- [Automatic Spatially-aware Fashion Concept Discovery](https://mlsanity.com/paper/1708.01311/) (2017)
- [Products-10K: A Large-scale Product Recognition Dataset](https://mlsanity.com/paper/2008.10545/) (2020)
- [The Met Dataset: Instance-level Recognition for Artworks](https://mlsanity.com/paper/2202.01747/) (2022)
- [RP2K: A Large-Scale Retail Product Dataset for Fine-Grained Image Classification](https://mlsanity.com/paper/2006.12634/) (2020)
- [Investigating the Vision Transformer Model for Image Retrieval Tasks](https://mlsanity.com/paper/2101.03771/) (2021)
- [Google Landmark Recognition 2020 Competition Third Place Solution](https://mlsanity.com/paper/2010.05350/) (2020)

## Built on by (most cited first, 1 of 1)

- [Towards Universal Image Embeddings: A Large-Scale Dataset and Challenge for Generic Image Representations](https://mlsanity.com/paper/2309.01858/) (2023)

## Abstract (arXiv)

This paper presents the 6th place solution to the Google Universal Image Embedding competition on Kaggle. Our approach is based on the CLIP architecture, a powerful pre-trained model used to learn visual representation from natural language supervision. We also utilized the SubCenter ArcFace loss with dynamic margins to improve the distinctive power of class separability and embeddings. Finally, a diverse dataset has been created based on the test's set categories and the leaderboard's feedback. By carefully crafting a training scheme to enhance transfer learning, our submission scored 0.685 on the private leaderboard.