Veryfi Lens Mobile Capture Framework first AI-based edge detection and document blur detection features for mobile web apps

With the new Veryfi Lens Mobile Capture Framework for Web, developers now have an easier way to transform document images and extract data through a combination of optical character recognition (OCR) and artificial intelligence (AI) for creating progressive web applications (PWA) built using common web technologies including HTML, CSS, JavaScript, and WebAssembly. PWAs work on any platform that has a standards-compliant browser across desktop and mobile devices.

The Veryfi technology delivers an enhanced user experience and improves accuracy compared with homegrown apps using built-in device cameras, plus much faster time to market.

Marketing Technology News: MarTech Interview with Lisa Arthur, Chief Marketing Officer at Sensedia

The Veryfi Lens Mobile Capture Framework for Web provides:

• Automatic edge detection: no need to move around the camera to “fit” into borders;
• Automatic document detection: recognizes receipts, invoices, and hotel folios;
• Auto-cropping: users do not have to manipulate captured images;
• Intelligent document pre-processing: removes skews, barrel, and pincushion distortions, and uses a proprietary compression algorithm to optimize image quality for available upload bandwidth;
• Automatic blur detection: returns a boolean value to indicate if a captured image is blurry or not, so that feedback can be provided to the user.

The release of Veryfi Lens Mobile Capture Framework marks the world’s first known AI-based edge detection and document blur detection features in a browser-based mobile app.

For companies looking to increase mobile user engagement and conversion rates, PWA apps are a compelling improvement over mobile website experiences. Now, companies no longer need to compromise on features or data extraction accuracy when providing PWA apps to their users. The average conversion increase from PWA apps is 36%, versus mobile websites and since the beginning of 2021, desktop PWA installs have grown by 270%. By switching to PWA apps, large companies across industries are seeing higher install rates, user engagement, conversion rates, and more.

“From the beginning, Veryfi understood the importance of high-quality image capture as a key driver in data extraction accuracy and enhanced user experience,” said Dmitry Birulia, co-founder of Veryfi. “With these recent innovations in Lens, we’re extending our lead in AI-driven document capture on mobile devices, delivering greater user engagement, better performance, and increased accuracy.”

Marketing Technology News: CDPs Still Critical For Marketers Despite ‘Data Clean Room Trend’

Now, the Veryfi Lens Mobile Capture Framework supports all mobile application technologies – native, wrapper, PWA, and web – so that developers can choose the appropriate technology for their needs and desired benefits. For many use cases, the software provider does not want to require users to download a full application through one of the application stores. Veryfi Lens Mobile Capture Framework provides advanced data capture functionality to improve the customer experience and data extraction quality, at scale, without requiring the download of a native application. Users can capture documents directly from a browser-based application, regardless of device type, and without requiring the user to install anything.

Veryfi provides the only mobile receipt capture technology that includes AI computer vision for receipt detection and image enhancement, as well as a long receipt capture mode. It helps transform unstructured documents into structured data – tapping into data that was previously unavailable for analysis.

By eliminating manual data entry, Veryfi enables organizations to accurately capture, extract and transform documents such as receipts, invoices, purchase orders, checks, credit cards, and W-9 forms into structured data, at scale. Veryfi uses advanced AI/ML technology, trained by hundreds of millions of documents over the past five years, to extract data and transform it into a structured format for 89 currencies, 39 languages, and over 110 defined fields such as vendor, total, bill to/ship to, purchase order and invoice numbers, any line item (product name, SKU, description), taxes, and more, which can then be accessed for a wide variety of business applications.