iYUNDONG‘s AI Automatically Extracts Sporting Event Highlights

iYUNDONG builds artificial intelligence (AI) tools that analyze media captured during sporting events to automatically extract key highlights. For example, by uploading a single photo, images of individual marathon runners can be automatically retrieved from a massive dataset that includes photos and videos from the event. iYUNDONG leverages Milvus, an open-source vector database built by Zilliz, to power its core image retrieval system.
Objective
From a single uploaded photo, accurately retrieve all images of a user from a massive, dynamic database.
Challenges
- Return accurate search results based on user uploaded photos that vary in resolution, size, clarity, angle, and other ways that complicate similarity search. - Make newly uploaded photos immediately available for search. - Run millisecond-level queries on a database of **60+ million** images. - Build a production ready application in **one month**.
Why verctor database
- Thanks to Milvus’ data flushing mechanism, iYUNDONG’s backend system is faster and more efficient because resource use is distributed across different nodes. - Support for real-time search on trillion-vector datasets gives users instantaneous results. - An active open-source community and powerful out-of-the-box features allowed iYUNDONG to operate on a tight development budget.
Results
- Accuracy rate of photo retrieval is constantly **above** **92%**. - Newly uploaded photos can be searched in real time. - Queries completed in milliseconds on average. - Project finished on time and under budget.
"Milvus’ performance is unparalleled. The platform allowed us to create a powerful, enterprise-grade image retrieval system on a short timeline with limited resources."
Read the full storyshare this
