DALLAS, Nov. 29, 2023 (GLOBE NEWSWIRE) — National banks and mortgage lenders are scrambling to incorporate image recognition into their appraisal review process, as the leading Government Sponsored Enterprise announced it will use “image recognition” to detect incorrect quality and condition ratings. Restb.ai, real estate’s leading AI-powered computer vision solutions provider, offers mortgage originators and appraisal management companies (AMCs) a GSE-compliant image validation solution with its computer vision technology.
GSEs require appraisal reports to include interior and exterior photos of subject properties. However, they only require one exterior front photo for each comparable. Recently, Fannie Mae analyzed more than a million appraisals using image recognition technology to compare appraisal reports to interior photos of comparables for comparing condition ratings. The differences in ratings were so significant that Fannie Mae is now using image recognition to detect condition errors.
With the GSE moves, mortgage lenders and appraisal firms seek to reduce condition errors by adopting computer vision technology into the appraisal process.
Recently, Fannie Mae noted image recognition technology was able to identify appraisal defects with 98% accuracy and significantly increase its efficiency, noting the new technology found “many defects that were previously impossible” for it to detect, adding that incorrect condition ratings can lead to missing or faulty adjustments to comparable sales, resulting in unsupported, inaccurate appraisals.
“When the GSEs talk, lenders and appraisers listen – and act,” said Tony Pistilli, General Manager, Valuations for Restb.ai, an appraisal industry veteran and respected valuation expert. “Image recognition is now a must-have. If Fannie Mae is doing it, appraisal providers and lenders need to be doing it too,” he added.
The significant shift by the secondary market to evaluate internal photos for quality and condition has resulted in rejecting appraisals when comparables are used without the appropriate adjustments.
“Lenders and AMCs can immediately benefit by reducing the number of appraisal corrections, mitigating loan repurchase risk, and improving appraisal turn times and increasing appraisal quality by adopting our computer vision solution,” said Nathan Brannen, Chief Product Officer for Restb.ai. “Computer vision can be a trusted source for quality and condition ratings, identifying property damage, home features not mentioned in the appraisal, and most importantly, protect the lender and appraisal provider by validating information in the appraisal,” he added.
In August, Restb.ai launched its Valuation Product Suite, which leverages its deep expertise in computer vision for real estate to offer an array of solutions designed to expedite the modernization of property appraisals.
Details about the Restb.ai computer vision solutions and its Valuation Product Suite are online here (restb.ai/customers/appraisals-inspections).
About Restb.ai
Restb.ai, the leader in AI-powered computer vision for real estate, provides image recognition and data enrichment solutions for many of the industry’s top brands and leading innovators. Its advanced AI-powered technology automatically analyzes property imagery to unlock visual insights at scale that empower real estate companies with relevant and actionable property intelligence. Restb.ai is like having a real estate expert instantly research and provide a deep insight into each of the 1 million property photos uploaded daily.
For more information on Restb.ai, visit its website. For Restb.ai-related media inquiries, please contact Maya Makarem at @restb.ai or [email protected] or Kevin Hawkins at 1-206-866-1220 or [email protected].
Media contacts:
Kevin Hawkins 1+ (206) 866-1220
[email protected] or
Maya Makarem
[email protected]
Photos accompanying this announcement are available at:
https://www.globenewswire.com/NewsRoom/AttachmentNg/e12029c2-84d2-4d23-9361-9f9b1e6b7c8e
https://www.globenewswire.com/NewsRoom/AttachmentNg/300c6db5-b4ef-4761-9d64-c3b315809229