Plainsight Blog

AI Week in Review #35 | 2022

September 2, 2022
Bennett Glace

Bennett Glace

This week in AI and Machine Learning: AI-powered drive-thrus, crackdowns on unregistered swimming pools, and more.

A Note from the Author

In a busy restaurant, service excellence is both an imperative and a challenge. Customers expect their favorite dishes fast and restaurant staff members are tasked with making various calculations on the fly to keep them satisfied. How long will the remaining servings of a popular entree last? When should the kitchen start cooking more? 

Plainsight recently helped a large North American restaurant chain answer these questions and address service challenges with computer vision. Through a partnership with World Wide Technology, Google Cloud, and Intel, the chain deployed custom-built Plainsight models to achieve 95% accuracy in estimating food quantities with overhead cameras. Learn more about the project and its additional benefits for the chain.

Artificial Intelligence News

Panera Tests Drive-Thru AI

COVID-19 restaurant closures contributed to a surge in demand for options other than on-location dining. QSR mainstays like McDonald’s like Yum! Foods stable of brands redoubled their emphasis on drive-thru service and restaurants of all types invested in service offerings like curbside pickup and delivery. 

Panera Bread is the latest brand to embrace AI for expediting order fulfillment and improving accuracy. Starting next week, two Upstate New York Panera locations will start taking drive-thru orders via a voice-based ordering system powered by AI. The chain has already experimented with using AI to monitor the temperature and volume of its coffee. 

Check out Plainsight’s webinar from early 2021 covering some of the ways computer vision models help generate customer insights to promote better customer service and help restaurants take advantage of increased demand for drive-thru and takeout dining.

A New Feature for DALL-E 2

DALL-E 2, the impressive AI-powered art generator, has taken the world by storm since its debut earlier this year. More than a million people are now testing out the feature for themselves. Creating remarkable images from simple prompts, the tool has wowed casual observers while raising questions about future use cases for AI-generated art and the future job security of artists in various fields. 

A new feature called outpainting will soon expand the canvas of DALL-E 2’s creation. In addition to presenting users with standard 1024 x 1024 images, the solution will now offer users the opportunity to flesh out their images in all directions while maintaining the same aspect ratio. Users already have access to inpainting, a feature enabling edits within specific regions of an image. 

Representatives from OpenAI, the team behind DALL-E 2, say outpainting was developed as a direct result of user feedback. Filmmakers and game designers leveraging the tool, for example, can use outpainting to envision scenes from new perspectives. Check out more details on the new feature – you’re sure to see it in action on social media soon.  

Detecting Unlicensed Swimming Pools with AI

With temperatures soaring across France, many citizens are looking to beat the heat by adding swimming pools to their properties. According to the nation’s financial authorities, however, a number of people have completed construction on the sly in hopes of avoiding property tax hikes. With the help of AI, regulators have identified more than 20,000 unreported pools across France throughout the last year. 

France’s government levies real estate taxes based on a property’s estimated rental value. So, a project like a pool is ultimately a mixed blessing for a homeowner. The tax office (Direction générale des Finances publiques) teamed up with software provider Capgemini to identify pools in aerial photographs. By cross-referencing visual data against property and tax databases, they were able to identify all those thousands of unregistered pools and line up potential millions in uncollected tax revenue. 

Read more about the project which should expand to cover the whole nation soon. 

Join our Community

See you next week! Until then, keep the conversation going on Plainsight’s AI Slack Channel

About the Author & Plainsight

Bennett Glace is a B2B technology content writer and cinephile from Philadelphia. He helps Plainsight in its mission to make vision AI accessible to entire enterprise teams.

Plainsight’s vision AI platform streamlines and optimizes the full computer vision lifecycle. From project strategy, through model deployment, and ongoing monitoring, Plainsight helps customers successfully create and operationalize vision AI applications to solve highly diverse business challenges.

Share