Welcome to Level Up! the show where we show you how to build solutions hands-on with Google Cloud Platform.
In this episode, Solutions Architect Markku Lepistö will show you how to analyze camera feeds with AI, using the Google Coral USB Accelerator. And then how to switch between cameras automatically, with both OBS Studio, and Blackmagic ATEM Mini Pro.
01:22 - Introduction to Google Coral USB Accelerator
01:32 - What is the Coral PoseNet Machine Learning model
02:06 - Creating a Coral Edge TPU client in Python code
02:13 - Inference response from PoseNet ML model
02:55 - Analyzing which camera the presenter is facing
03:36 - Switching cameras programmatically with OBS Studio
05:06 - Controlling OBS scene switching with AI
06:01 - Blackmagic ATEM Mini and atemOSC
06:36 - Controlling ATEM Mini camera switching with AI
Free Trial: Google Cloud Platform → [ Ссылка ]
Hands-on training with Qwiklabs
→ Application Development - Python - [ Ссылка ]
→ Machine Learning APIs - [ Ссылка ]
→ Google Cloud Platform Essentials - [ Ссылка ]
Follow us on Twitter
→ [ Ссылка ]
→ [ Ссылка ]
→ [ Ссылка ]
Subscribe to our Google Cloud APAC channel for more for more episodes where we dive deeper and build on previous examples → [ Ссылка ]
References:
PersonLab: Person Pose Estimation and Instance Segmentation with a Bottom-Up, Part-Based, Geometric Embedding Model. George Papandreou, Tyler Lixuan Zhu et al
[ Ссылка ]
Coral PoseNet. Mike Tyka et al
[ Ссылка ]
