Applied ML: A sports analytics & contract intelligence use case

GDG Cloud Netherlands
Tue, Feb 18, 2020, 5:30 PM (CET)

2 RSVP'ed

About this event

Abstract talk 1: Video Analytics for Football games

This ML use case is about how we used Google Dataflow to analyze and process, in near-real time, football game stream feeds. The goal was to determine events (start of game, team detection, player tracking, ball tracking) and performing analytics on these videos (duration, ball possession, score, ..) with ML models.

In particular, Apache Beam was used with Python SDK to create streaming pipelines. In this pipeline we used sliding windows to chunk the video frames into sets of sequence of frames, which could be used as input for different machine learning models.

Our machine learning models are hosted on GPUs via TF-serving on Kubernetes. The output of the machine learning models were written to Google Cloud Bigtable, which was used by a front-end player to visualize these features on top of the stream feed (with a small delay).

Other components used: Google Cloud Pub/Sub, Google Kubernetes Engine, Google Cloud Bigtable


Abstract talk 2: Contract intelligence to automate knowledge extraction from risk prone documents

Data is the new gold. But.. How can you unlock the power of your data efficiently? In this talk, we reflect on our application that uses the power of Machine Learning to automate knowledge extraction from pdf documents.

More specifically, we apply our models on housing deeds. The result is that less manual work is required from notaries and results are being gathered on a bigger scale. On top of that, we have made a pipeline to pseudomys the data in order to mitigate privacy risks.

Google components used: Pub/Sub, GKE, Google Dataflow, Google Bigquery, AI platform


Intro by Google (5 mins)

Talk 1: Video Analytics for Football games (45 min)

Break (15 min)

Talk 2: Contract intelligence to automate knowledge extraction from risk prone documents (45 min)

Quiz + price (10 min)

Networking + food

Speaker bio’s:

Brecht Coghe:

Brecht is a Senior Machine Learning Engineer at ML6. He has been working at ML6 for 3 years now and has been involved in multiple big-scale machine learning projects. Prior to ML6, Brecht was a PhD candidate in the field of bioinformatics.

Stef Ruinard:

Stef Ruinard is a Machine Learning Engineer at ML6. As an AI-enthusiast with a special focus on applied machine learning, he is devoted to empowering organisations with AI. He greatly enjoys working in the startup culture and has a special interest in Reinforcement Learning and Robotics. If he is not coding, you'll most likely find him reading business books or cycling through a, random, forest.