Unlocking the next stage in computer vision with deep neural networks

Description

One major end goal of artificial intelligence and deep learning technologies has always been to mimic how the human brain processes and interprets data through the senses. While much work remains to realize that goal, we’re starting to see real and meaningful progress toward this vision. These tools now enable machines to recreate how the human brain interprets images—essentially giving computers eyes and the ability to see. This concept has major implications for businesses that need to extract meaning from overwhelming quantities of unstructured data, like digital images.

Data scientists at Zillow Group are leveraging computer vision-based in deep neural networks to test new ways to estimate a home’s value based on specific image criteria. They’re training their solutions to interpret home attributes—from hardwood floors to granite countertops—to help infer the value of a home.

This technology is in its infancy. The organizations that use it most effectively will be those who know which questions to ask and what challenges to solve first. Josh walks you through the core problems within deep neural networks that his team is working to solve and how organizations can begin to experiment with its capabilities.