IoT For All Podcast · 24 May 2021 · 30m
Vishal Singh, Founder and CTO of ZenduIT, discusses the evolution of IoT from simple GPS tracking to complex computer vision and AI solutions for fleet management and airport operations. He explains how ZenduIT uses edge computing to detect driver fatigue and phone usage, emphasizing the critical role of high-quality data and proper camera positioning in building reliable AI models. Singh also addresses the challenges of AI accuracy, the security risks of connected cameras, and the future trend of managed services for video analytics.
Where this conversation describes AI being used to build or deliver a product. Part of the AI in practice tracker.
Engineering & simulation
The team develops an AI model to identify drivers by matching their faces against a database, requiring careful camera positioning and large datasets to handle different angles.
“what we're trying to work on is like saying, hey, let's let's look at who's coming, who's in the vehicle by their face. And let's match them up. … the big challenge is positioning where the camera is and having enough data to accurately determine that this is John.”
Speaker 1 · Hear it at 14:46
companies scaling up
Other shows on the same ground, closest first.
Ryan Prosser and Bethany Ibarra of Very interview leaders at the 2023 AIMST Conference, including Solomon Levy of Roboflow, Andrew Hines of Stratus Technologies, Austin Byers of TX1, Suzanne Loda of Agco, and Ryan Legg of Iconics. They discuss the shift toward operational technology (OT) security, the use of computer vision for quality assurance, and the challenges of integrating IT and OT systems in industrial settings. Key decisions involve prioritizing availability over confidentiality in OT, consolidating edge computing workloads, and simplifying deployment for legacy manufacturing environments.
Balint Horvath shares insights from the IoT Tech Expo Europe, highlighting the rise of AI and IoT investments and the need for realistic expectations, citing companies like Aiir Innovations, Asgard, and Circuit. He discusses specific applications including predictive maintenance, livestock monitoring, and global connectivity solutions like Keep Go and Hyper. The episode features an interview with Cassandra Lore of Aiir Innovations, detailing their computer vision system for automated airplane engine inspections. Horvath emphasizes the strategic approach of assisting inspectors before moving to full automation, supported by partnerships with KLM and cloud computing resources.
Evan Welbourne, Head of AI and Data at Samsara, discusses the evolution of IoT from a distinct technology to an invisible infrastructure, emphasizing that the real value lies in data analytics and edge computing. He details Samsara's pivot from a temperature sensor to a vehicle gateway, highlighting how thermal constraints in AI dash cams drive hardware design decisions. The conversation explores the future of neuromorphic computing, data ownership in connected devices, and the rise of edge AI for real-time safety insights.
Tanuj Thapliyal, Co-Founder and CEO of Spot AI, explains how his company transforms industrial video surveillance into operational intelligence for manufacturers. He details the origin of Spot AI from Stanford, its focus on empowering frontline workers with easy-to-use video data, and the company's recent $20 million Series A funding. Thapliyal also discusses critical cybersecurity architectures, including zero-trust models and local storage, and shares examples of how video intelligence solves real-world manufacturing problems like conveyor belt failures.
Jonathan Berte, Founder of RoboVision, explains how his company uses AI-driven machine vision to solve the labor shortage and complexity challenges in modern manufacturing. He details the shift from manual algorithm creation to automated deep learning, enabling scalable quality control and object detection for clients like Audi. Berte also discusses the strategic trade-offs between cloud and edge computing for industrial applications and the role of flexible robotics in replacing tedious human tasks.
The summary, chapters and topics are generated from the episode’s audio by this directory, not written by IoT For All Podcast. The episode itself belongs to its publisher.