Web-Scale Generic Object Detection at Microsoft Bing
Stephen Xi Chen, Saurajit Mukherjee, Unmesh Phadke, Tingting Wang,, Junwon Park, Ravi Theja Yada

TL;DR
This paper introduces GenOD, a large-scale object detection system for Bing that detects over 900 categories in real-time, improving search relevance and user engagement through efficient data collection, training, and deployment.
Contribution
The paper presents a novel large-scale object detection system with a scalable data pipeline, improved update efficiency, and significant benefits for visual search applications.
Findings
Over 20% improvement in weighted average precision.
81.5% reduction in labeling cost.
54.9% increase in search relevance.
Abstract
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…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
Methodstravel james
