Multi - camera tracking is the practice of tracking objects and accurately measuring activities across multiple cameras in large spaces (such as factories, retail environments, and public areas that a single camera can’t cover), which enables effective monitoring and management of these spaces.
→ View more
Retail product shrinkage hits $100B yearly (over 65% from rising theft). NVIDIA’s AI workflow, using Metropolis microservices and pretrained models, quickly builds anti-theft apps, leveraging few-shot learning to recognize vast products and send actionable alerts.
→ View more
In the digital age, as consumers shift to e - commerce, they demand more personalized shopping experiences and extensive product choices, and will switch brands after poor experiences. Facing complex inventory and fierce competition, retailers leverage AI and data science to provide real - time, hyper - personalized experiences, so as to attract customers, increase cart size, and improve conversion.
→ View more
NVIDIA leverages Omniverse (powered by OpenUSD), NVIDIA AI, and technologies like RTX/Isaac, plus user-friendly developer tools, to enable virtual factory solutions—supporting unified data pipelines, physically accurate simulation, AI-driven optimization, and seamless collaboration.
→ View more
Training real datasets for AI models are time-consuming and expensive to collect and label, hindering development; however, synthetic data generated via computer simulation can be combined with real data to train computer vision AI models, significantly saving training time and costs.
→ View more