Live Sport Analysis
How to keep 10+ different datasets organized, manage your annotation workflows smoothly, and bring together 10 Computer Vision engineers and 10+ labelers.
SPORT
COMPUTER VISION ENGINEER
SCALE PRODUCTION
Context
A French AI scaleup specialized in sports analytics is rapidly scaling its AI engineering team and processes. This company offers a real-time football analytics platform able to provide insights to coaches and scouts.
The team is very knowledgeable in AI and computer vision (CV) and they are looking for powerful tools to help the team be as efficient as possible.
One
1 single point of truth for every CV engineers.
+33%
Additional time spent by data scientists on high-value tasks.
2 month
From 6 to 2 months to deliver analytics for a new sport.
Datalake: a centralized datalake to index, search and store all the data and track sources and metadata.
Monitoring System: a dashboard which collects CV-oriented metrics to assess the relevance of new live data.
Model Registry: a single point of truth to version and centralize all models.
Continuous Training: a user-friendly interface for experts to review predictions, and reviewed images to automatically update new training datasets.
