Rapid Growth of Computer Vision Improves Experiences and Operations Across Industries
In this article
The rapid growth and availability of generative artificial intelligence (GenAI) has been a major topic in 2023 with volumes being written about its emergence and much more to come as AI continues its exponential growth and adoption.
However, artificial intelligence (AI) has made significant advances in more areas than just GenAI. One such area of growth is AI's increasing role across a breadth of industries in the evermore sophisticated world of computer vision applications. In fact, a 2022 Research and Markets report predicted the global computer vision market was expected to grow from $13.45 billion to $14.95 billion in 2023 — a compound annual growth rate (CAGR) of 11.2 percent.
Recognizing the significance of visualization to emerging AI solutions, WWT and longtime ally Intel® co-hosted a Computer Vision Summit in August 2023 at WWT's headquarters in St. Louis. Industry executives and other leaders gathered at the Summit to share insights, trends, challenges and opportunities. Also on hand were representatives from across the computer vision ecosystem — spanning hardware, software, infrastructure, cloud and other companies — for knowledge-sharing to help scale growth.
Among the attendees was another WWT ally, WaitTime, a pioneer in computer vision technology and the industry leader in crowd intelligence. Their unique solution is optimized on Intel technology and has applications across a range of industries and organizations providing improvements in customer and employee experiences as well as operations.
A number of factors have contributed to computer vision's expanding role in advancing AI-powered solutions, including:
- Safety considerations: For example, in potentially hazardous environments such as mining, Computer Vision can operate in conditions that avoid placing human monitors at risk.
- Growth of digital channels: Today, almost everyone carries a smartphone or portable computing device with a range of productivity apps. Computer vision can help provide these users with a richer experience using real-time data.
- Increasing labor costs: As computer vision capabilities improve and employee payrolls rise, technology becomes a more cost-effective alternative to staffing up, providing a better return on investment.
- Edge capabilities without costly GPUs: Previously, computer vision would have required a significant investment in additional graphics processing. Today's edge computer vision solutions don't need special GPUs, making them a more viable, affordable option.
- Wider, less expensive hardware choices: Cameras used to be a significant barrier to entry due to their cost and proprietary nature; now there is a greater selection available with more capabilities at a lower cost.
- Higher data accuracy: Certain solution providers that employ beacons and Wi-Fi pinging used to make dubious claims regarding their data accuracy. Leveraging edge computing in advancing AI and GenAI for businesses is pivotal. For computer vision, edge solutions facilitate real-time or near-real-time data analysis. This simplifies the process of training AI models and enhancing the performance of AI-driven applications like WaitTime, ensuring a measurably high degree of accuracy to make data more valuable and actionable for every use.
Owing to a host of dynamics like these, computer vision has become a necessary component for AI-powered use cases across industries. One example is the ability to glean fast, actionable insights from crowd intelligence using the various WaitTime, Intel and edge solutions enabled through WWT.
Established nearly 10 years ago, WaitTime was one of the first companies to enter the nascent computer vision market, focusing on developing solutions for gleaning crowd intelligence data in real time. This early experience gave WaitTime an advantage when it came to applying its technology to a host of new environments.
WaitTime stands out for its state-of-the-art, specialized and patented software that leverages camera data to provide the insights needed to optimize operations, improve crowd management, and enhance interactions for customers and employees alike.
A key differentiator for WaitTime is its four proprietary algorithms for evaluating and controlling crowds:
- Queuing: Analyzing patterns and behaviors of how people form lines organically in various locations (e.g., at a restroom or retail checkout queue).
- Massing: Assessing crowd density and traffic movement in real-time, useful for venues like concessions, cafeterias and sports betting.
- Stanchion: Designed to analyze artificial or tightly organized "switchback" queuing strategies in transportation terminals, events and other funnel points.
- Entry/exit: Measuring how traffic finds its way into, around and out of the environment; useful for entertainment venues, office buildings and seating areas.
WaitTime is designed for optimum performance on the latest Intel technology, including 4th Gen Intel Xeon® Scalable processors for high-demand workloads, and a standards-based oneAPI programming model to support disparate hardware architectures and vendors.
The WaitTime and Intel solution is not only unique and highly performant, but its output of accurate, detailed data can be integrated with a diverse range of tools for use in point-of-sale, marketing, weather forecasting, event operations and much more.
The WaitTime, Intel and WWT real-time crowd intelligence solutions can be employed in a wide range of applications and environments, with results that:
- Increase customer satisfaction: For example, a computer vision application might be designed to delight guests attending an event with helpful information such as wait times for restroom queues, food vendors, merchants and other destinations — providing a time-saving and frictionless experience. Monitoring wait times can also ease congestion in public-sector buildings and travel terminals.
- Monetize foot traffic: In mall retail environments and event venues, management can use computer vision to identify high-traffic areas to justify tiered space-leasing terms with higher rents to match the premium location value.
- Enhance safety and security: Monitoring crowds in sports and entertainment venues with computer vision can alert personnel to security issues. The technology can also track suspicious activity and incursions into off-limits areas to prevent mischief.
In addition, computer vision can be utilized to:
- Improve staff productivity: In busy restaurant environments, computer vision can minimize employee stress over when to replenish popular menu items, making it easier to fulfill anticipated customer needs.
- Reduce waste: Similarly, computer vision use in commercial kitchens can ensure the optimum level of inventory is on hand to satisfy customer demand while minimizing end-of-day leftovers.
- Enable predictive maintenance: In industrial operations, computer vision can monitor gears for wear and tear, indicating when service or replacement should be scheduled for non-peak hours.
These are but a few of the ways the real-time crowd intelligence solution from WaitTime and Intel can contribute to AI-driven benefits across a host of industries, enriching both the customer and employee experience while reducing costs and promoting efficient operations. WWT can help assess how computer vision can play a role in improving your organization.
Computer vision offers significant value for any organization that can leverage accurate data about the real-time status of people and things in motion. With more than 10 years of intense AI research and development work, plus our experience collaborating with industry leaders such as WaitTime and Intel, WWT can accelerate your adoption of computer vision solutions.
We can help you:
- Identify the challenges and use cases computer vision can address.
- Implement solutions that integrate technology leaders across domains, including hardware, software, security, data strategy and management and more.
- Compare and test use-case solutions in our Advanced Technology Center (ATC), WWT's multi-campus R&D ecosystem that is available in person and virtually.
- Develop proofs of concept, ROI calculations and points of view.
- Perform ongoing support to sustain and extend your computer vision applications.
WWT brings together all elements needed to deliver computer vision at scale, whether across your network or in the cloud. Through our network of global integration centers, we can handle every step of the lifecycle — from solution acquisition, testing and bundling to shipment, deployment and integration.