6th Place Solution to Google Universal Image Embedding
This is an incremental improvement for image embedding tasks, specifically targeting competition performance.
The authors tackled the Google Universal Image Embedding competition by using a CLIP-based approach with SubCenter ArcFace loss and a custom dataset, achieving a score of 0.685 on the private leaderboard.
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.