LEONID STARIKOV || HOW AI WORKS IN MANUFACTURING TODAY || VMS PLATFORM
#neuralnetworks #automation
In our first episode of the corporate podcast, Leonid Starikov, founder of "Nedremlyushchee Oko," discusses the challenges of automating business and production using neural networks. We discuss how modern artificial intelligence technologies are transforming industrial enterprises and retail, and what prospects the development of an innovative VMS (Video Management System) platform offers for the widespread adoption of neural network solutions.
Leonid Starikov discusses the company's journey from the first successful implementations of video-based object recognition systems in 2019, such as automating checkpoints at factories, to developing complex neural networks for unique client needs. The podcast discusses practical examples of neural network use:
Optimization of technological processes in metallurgy (for example, monitoring the correct handling of hot ingots by a tong crane to prevent defects).
Improving efficiency in retail (analysis of product display, queue monitoring, heat maps of customer behavior).
Automation of quality control and product counting on conveyor belts (counting bricks, barrels, and train cars).
Logistics solutions (monitoring the time fruit pallets spend in the cold, monitoring available parking spaces).
You will learn how neural networks are trained for specific production needs, including the stages of data collection and labeling, the timeframes, and estimated cost of developing custom neural network detectors. Leonid explains why the successful implementation of AI in industry requires a synergy of IT competencies, engineering experience, and video surveillance system installation skills.
The key topic of the discussion was the development of a VMS platform from "Nedremlyushchee Oko." This no-code/low-code tool will allow engineers at enterprises to independently create and configure automation scripts based on neural networks, without advanced programming skills. The discussion explores how such a platform can democratize access to video analytics technologies and accelerate the digital transformation of industry. Leonid Starikov shares his vision of a future where neural networks will become an integral part of production processes, and why major tech giants aren't yet focused on this market segment.
If you're interested in how neural networks can improve your business's efficiency, what ready-made solutions exist, and how to begin implementing artificial intelligence at your enterprise, watch this episode.
Contents:
00:00 Introduction
01:30 When "Needless Eye" Implemented Its First Neural Network
03:14 Why Is Neural Network Implementation So Slow?
7:37 What Problems Are We Solving with Video Cameras and Neural Networks
14:03 Why Are Some Neural Networks More Expensive Than Others
17:28 How Neural Networks Are Trained
20:15 Why There Are Almost No Industrial Neural Network Manufacturers in Russia
23:35 What Neural Networks Did "Needless Eye" Create and Sell?
23:53 Combining Different Processes in a Single Script
29:30 Why Is It Better to Implement Neural Networks in Old Plants
31:08 VMS Platform: Automating Processes Without Code or Programmers
35:19 Why Open AI and Yandex Don't Do This
Regular news about neural networks, automation, and the company's successes: https://t.me/nedroko