In this paper, we introduce a web-scale general visual search system deployed in Microsoft Bing. The system accommodates tens of billions of images in the index, with thousands of features for each image, and can respond in less than 200 ms. In order to overcome the challenges in relevance, latency, and scalability in such large scale of data, we employ a cascaded learning-to-rank framework based on various latest deep learning visual features, and deploy in a distributed heterogeneous computing platform. Quantitative and qualitative experiments show that our system is able to support various applications on Bing website and apps.
@article{arxiv.1802.04914,
title = {Web-Scale Responsive Visual Search at Bing},
author = {Houdong Hu and Yan Wang and Linjun Yang and Pavel Komlev and Li Huang and Xi Chen and Jiapei Huang and Ye Wu and Meenaz Merchant and Arun Sacheti},
journal= {arXiv preprint arXiv:1802.04914},
year = {2018}
}