We kicked off September with another insightful AI in Action Virtual meetup as we were joined by three excellent speakers from all over the globe to share their fascinating work into the world of Computer Vision.
Justin Shenk, Co-Founder & CTO at VisioLab kicked off proceedings by taking us through his insights into Learning with limited labeled data. Based in Berlin, VisioLab brings vision-based food recognition to canteens to enable fast, scalable checkouts.
To make usage simple for canteen owners, dishes have to be recognized with only a few labelled images, far less than the thousands typically used for image recognition. Justin spoke about how VisioLab addresses these challenges from data augmentation strategies and active learning with customer feedback to models trained in the cloud and deployed on the edge.
Next up Ganesh Sistu who is Computer Vision and Deep Learning Expert at Valeo Vision Systems shared some fascinating applications of the work they do within autonomous driving. Deep Learning has revolutionized vehicle’s perception abilities and made the automated cars real.
Ganesh spoke about how Deep Learning is playing a crucial role in autonomous driving applications like object detection, semantic segmentation, motion estimation and camera lens disturbance detection. He also spoke about how they build fisheye camera-based visual perception solutions for automated parking systems and present a complex life cycle of deep learning projects.
Last but not least, Ben Taylor, who is a Chief AI Evangelist at DataRobot in the United States talked about Teaching CV systems to learn and comprehend like a human. As AI workloads have expanded beyond experimental notebooks, DataRobot has developed data ingest and MLOps to ensure trust and reliability for the enterprise.
Ben explored some of the storytelling methods that work for deep learning and discussed their work in Computer Vision and the latest Deep Learning techniques. He also shared his opinions on the concept of intelligence and a better Turing test for machines that finally think.
Watch this fascinating meetup in full below.