通过大规模实验，总结图像检索训练的最佳实践。
All You Need to Know About Training Image Retrieval Models
- 系统性测试多种训练因素对检索效果的影响。
- 在多个数据集上验证了通用最优配置。
- 适合想提升图像检索性能的研究者和工程师。
图像检索旨在从数据库中找到与查询图像最相似的图像。检索系统的性能受诸多训练因素影响，包括嵌入模型架构、损失函数、数据采样策略、挖掘函数、学习率和批量大小。本文通过运行数万次训练实验，系统分析了这些因素对检索准确率的影响，并发现跨多个数据集均适用的最佳实践。相关代码已公开于 https://github.com/gmberton/image-retrieval。
原文摘要 · Abstract (English)
Image retrieval is the task of finding images in a database that are most similar to a given query image. The performance of an image retrieval pipeline depends on many training-time factors, including the embedding model architecture, loss function, data sampler, mining function, learning rate(s), and batch size. In this work, we run tens of thousands of training runs to understand the effect each of these factors has on retrieval accuracy. We also discover best practices that hold across multiple datasets. The code is available at https://github.com/gmberton/image-retrieval
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。