Yimin Yang: Reimagining the house-hunting experience with AI


Yimin Yang, a professor in the department of electrical and computer engineering at Western University. (Jacob Arts/Western Engineering photo)

Western Engineering researchers have developed a new generative Artificial Intelligence (GenAI) model for Wahi that has the potential to change the way people find their perfect home.  

Yimin Yang, a professor in the department of electrical and computer engineering, is leading a team that is training an AI model that can take tens of thousands of real estate property photos and accurately identify living spaces such as kitchens, bathrooms, dining rooms and living rooms.  

He aims to demonstrate the application of artificial intelligence in real estate, specifically in image recognition and annotation, can offer a more personalized and efficient search experience for homebuyers.
This research is part of a collaboration with Dr. Eman Nejad and his Data Science team in Wahi, a Canadian real estate agency. Yang and Wahi envision this AI model helping home buyers find the exact living space that’s right for them.  

The key features of GenAI, in addition to accurately identifying living spaces in property photos, include categorizing those spaces based on renovation factors and enabling advanced image-based filters, providing users with a powerful tool to search for homes based on their preferences.


The GenAI can accurately discern between different living spaces, such as a dining room or kitchen, as well as the style of living space. (photos provided)

This shift from traditional functional filters to more aesthetic and materials-focused filters adds a new dimension to the home search process, allowing consumers to find properties more quickly and accurately. 

“Whether a user is in search of a sleek, modern kitchen equipped with the latest amenities, or a warm, inviting living room that exudes traditional charm, our system provides invaluable insights and indicators to guide them through their property journey,” says Yang.  

He believes the emphasis on photos as the most important aspect of real estate listings aligns with consumer preferences, and the ability to filter based on specific characteristics like "renovated kitchen" or "finished basement" enhances the user experience. This not only simplifies the search process but also ensures that users are presented with listings that match their unique preferences.

Wahi's adoption of GenAI sets it apart from other real estate platforms, says Yang, offering a competitive edge by providing a more tailored and user-friendly experience. He believes the integration of artificial intelligence in the real estate industry is a testament to the continuous evolution of technology to meet the changing needs and expectations of consumers. This collaboration exemplifies how interdisciplinary partnerships can lead to innovative solutions that redefine traditional processes.