Web-Scale Generic Object Detection at Microsoft Bing
- Stephen Xi Chen ,
- Saurajit Mukherjee ,
- Unmesh Phadke ,
- Tingting Wang ,
- Junwon Park ,
- Ravi Theja Yada ,
- Xi Chen
KDD 2021 |
In this paper, we present Generic Object Detection (GenOD), one of the largest object detection systems deployed to a web-scale general visual search engine that can detect over 900 categories for all Microsoft Bing Visual Search queries in near real-time. It acts as a fundamental visual query understanding service that provides object-centric information and shows gains in multiple production scenarios, improving upon domain-specific models. We discuss the challenges of collecting data, training, deploying and updating such a large-scale object detection model with multiple dependencies. We discuss a data collection pipeline that reduces per-bounding box labeling cost by 81.5% and latency by 61.2% while improving on annotation quality. We show that GenOD can improve weighted average precision by over 20% compared to multiple domain-specific models. We also improve the model update agility by nearly 2 times with the proposed disjoint detector training compared to joint fine-tuning. Finally we demonstrate how GenOD benefits visual search applications by significantly improving object-level search relevance by 54.9% and user engagement by 59.9%.