New AI-driven digital twin models are changing the foundations of retail operations and product development
CINCINNATI (PRWEB) April 08, 2021
NVIDIA'S GPU Technology Conference will showcase advances in hardware and software, including two sessions by Kinetic Vision on accelerating the power of Artificial Intelligence (AI). The conference starts on Monday, April 12th and is virtual. Registration is at NVIDIA.com.
“The sessions at NVIDIA’S GTC event are always illuminating. Kinetic Vision’s two presentations will show how new AI-driven digital twin models are changing the foundations of retail operations and product development,” said Jeremy Jarrett, Executive Vice President of Kinetic Vision.
The first session is a deep dive into physics-informed neural networks (PINNs). This innovative technology is able to provide real-time simulation results within the design environment, and could even create 'intelligent CAD' to guide the user towards the most functional design. The benefit of a physics informed neural network is the ability to perform bi-directional simulation, to scale linearly with additional GPUs, and to utilize an unstructured point cloud to enable meshless implementations. These advances can greatly improve the efficacy and speed of the product development process. The session is entitled “Using Physics Informed Neural Networks and SimNet to Accelerate Product Development” (Session ID: S31740)
The second session tackles the advanced AI techniques required to optimize retail inventory and pick-and-pack operations. Kinetic Vision and NVIDIA will show how a true digital twin can be constructed utilizing an automated synthetic data pipeline with AIVision® that works in concert with Intelligent Video Analytics (IVA). The benefit of this combination of a synthetic data pipeline with IVA is a highly scalable system where new products and new product variations in the retail operational model are easily accommodated. The session is entitled “Novel Approach to Deploy Highly Accurate AI Retail Computer Vision Applications at the Edge” (Session ID: S31538).