We are delighted to announce the Eight Computer Vision in Production event. This event is going to be focused onTransforming technologies: sports, geospatial, fashion and Machine Learning on Edge. Thanks to our wonderful host, Planet. for providing us with a location & refreshments for this event.
We have four Excellent talks lined up, before the talks we will have beers, pizza and networking for 15 minutes and we commence the talks at 6:30pm.
(6:30pm) Anthony Kelly - Host, Trust in SODA
(6:35pm) Michael & Neha - CTO & Deep Learning, AI Golf
(7:00pm) Hicham Badri - Senior Computer Vision Scientist & Co-Founder, Mobius Labs
(7:30pm) Alexander Seibert, Head of Machine Learning and Computer Vision, ID Vision
(7:55pm) Sara Bahloul - Senior Manager of Imaging Operations, Planet
Speaker Bio & topic:
Michael & Neha
The challenges of GolfAI sport IoT device which features computer Vision, AI, DL, Big data, statistic, battery management, App, gamification. We are adding gamification to real world golf training games. Agile methods for rapid prototyping, like 3D print. (additive manufacturing)
Dr. Hicham Badri is a Co-Founder and Senior Research Scientist at Mobius Labs GmbH. He holds a PhD in Computer Vision from Université de Bordeaux/INRIA. Prior to Mobius Labs, he was doing R&D at EyeEm. His research topics include on-device machine learning and few-shot learning.
The talk is about machine learning on edge devices (mobile phones, embedded boards, etc.) : what are the challenges, the tricks and the things to watch out for before deploying models that run on edge devices.
Speaker Subject: Player number classification
Speaker Subject: A satellite image data provider's perspective on Computer Vision for image processing and image quality assessment
Very early on in the imaging chain, when the image data just made it from on board the satellite into the ground processing systems, there is a whole suite of computer vision techniques that is utilized before the data is fully processed and released. We'll see some of those validation and processing steps, which ensure good and uniform data quality, serve to monitor the health of the cameras in space, while preserving full data integrity and credibility. The objective is to create the fundamental dataset to enable any further exploitation and information extraction down the line.