How to Identify Multifamily Value-Add Deals with Computer Vision

Published by
Marc Rutzen
on
October 25, 2023
How to Identify Multifamily Value-Add Deals with Computer Vision

Finding the Worst House on the Best Block at Scale

A few months ago, when we first started developing computer vision algorithms to assess real estate condition and quality, we were approached by a single-family value-add developer who wanted to use our real estate image extraction API to find the "worst house on the best block" at scale. He ran our QualityScore API over most of the city of Philadelphia, processing millions of interior and exterior listing photos to identify mismanaged properties or those in disrepair, and actually found a few good deals from the effort.

This process also helped us realize that we needed to add appraisal and site inspection photos to our dataset - listing photos are just too perfect. With most of our training set comprised of photos with ideal staging, lighting and angles, this customer showed that we were unfairly penalizing photos with poor lighting or staging. Fortunately, we were able to add hundreds of thousands of labeled appraisal images to the training set, balancing out our algorithms to detect underlying real estate quality vs image quality.

The use case of identifying single and multifamily value-add deals at scale is one that resonates with us, and we have been working on a way to analyze market rents and photos at the same time to identify deals with below market rents and poor condition. In this post, we talk about what computer vision is and how it can be used in real estate.

What is Computer Vision?

Computer vision is a field of artificial intelligence (AI) and computer science that focuses on enabling computers to interpret and make decisions based on visual data, such as images and videos, in a way that is similar to human visual perception. The ultimate goal of computer vision is to teach machines to "see" and interpret the visual world as humans do, and even beyond human capabilities in certain aspects.

Key concepts and applications of computer vision include:

  1. Image Recognition: This is the capability of software to identify objects, places, people, writing, and actions in images. Social media platforms use image recognition to tag and categorize photos, for example.
  2. Object Detection: While image recognition identifies what's in an image, object detection goes further by locating specific objects within the image. This is often represented by bounding boxes around the identified objects.
  3. Image Segmentation: This process involves partitioning an image into multiple segments or "pixels" to simplify or change the representation of an image into something more meaningful, making it easier to analyze.
  4. Face Recognition: This is a specialized form of image recognition that identifies and verifies individuals based on their facial features. It's widely used in security systems and various apps.
  5. Motion Analysis: Computer vision can track and analyze the movement of objects, which is crucial in video surveillance, traffic monitoring, and sports analysis.
  6. Scene Reconstruction: Creating a 3D model of a scene from multiple images or videos is another application of computer vision, often used in the realm of augmented reality or film production.
  7. Augmented Reality (AR): AR overlays virtual objects onto the real world, and computer vision is essential for understanding the environment and ensuring that virtual elements are positioned and scaled correctly.
  8. Autonomous Vehicles: Cars and drones that can navigate the world on their own rely heavily on computer vision to understand their surroundings, detect obstacles, and make real-time decisions.
  9. Medical Image Analysis: In the medical field, computer vision is used to analyze images (like X-rays, MRIs, and CT scans) to assist doctors in diagnosis and treatment planning.

To achieve these tasks, computer vision employs various algorithms and techniques that often involve deep learning, a subset of machine learning, especially convolutional neural networks (CNNs) and Transformer (ViT). These neural networks are trained using vast datasets of labeled images, enabling them to recognize and interpret new, unseen images. Essentially, computer vision seeks to replicate the intricacies of human vision and perception but also aims to surpass human capabilities by extracting insights and details that might be missed by the human eye.

How Can Computer Vision be used in Real Estate?

Using computer vision to identify value-add deals in the multifamily real estate sector can be a game-changer. Value-add deals refer to properties that offer the potential for increased returns through various enhancements, such as physical improvements, operational changes, or market position improvements. Computer vision, when applied correctly, can automate and enhance many tasks associated with spotting such opportunities. Here's how:

  1. Property Condition Analysis: By analyzing exterior and interior images of multifamily properties, computer vision algorithms can identify signs of wear, aging, or damage. Detecting properties that require cosmetic or structural upgrades can help investors pinpoint potential value-add opportunities.
  2. Amenities Assessment: Computer vision can be used to scan property images to identify the presence or absence of amenities (like swimming pools, fitness centers, communal areas). Properties lacking certain amenities in markets where they're standard might present value-add opportunities.
  3. Comparative Analysis: By comparing images of a target property with those of higher-end or renovated properties in the same market, computer vision can help identify visual disparities. This can highlight potential areas of improvement.
  4. Land Utilization: Aerial images, often sourced from drones, can be analyzed to assess land usage. Computer vision can identify underutilized spaces that might be converted into additional amenities, parking, or even additional housing units.
  5. Trend Analysis: Over time, computer vision can help analyze visual data from various properties to identify design and amenity trends in the market. Recognizing what's becoming popular can guide value-add decisions.
  6. Neighborhood Analysis: Using street-view images or satellite imagery, computer vision can gauge the general condition and development level of neighborhoods. Identifying up-and-coming neighborhoods can spotlight properties that may benefit from broader area improvements.
  7. Facial Analysis for Demographics: While treaded carefully due to privacy concerns, facial analysis can, in theory, determine the general age, mood, or demographic makeup of residents or visitors to an area. This could guide investors on the potential need for community-specific amenities or adjustments.
  8. Foot Traffic Analysis: For multifamily properties with commercial components or those considering the addition of such components (like ground-floor retail), computer vision can analyze foot traffic patterns to determine the viability of commercial ventures.
  9. Historical Assessment: In areas with historical buildings, computer vision can help identify architectural elements that need preservation. Restoring and highlighting these features can add unique value.
  10. Safety and Security: Analyzing images or videos of properties to identify potential security vulnerabilities (like poorly lit areas or easily accessible windows) can spotlight safety-enhancing value-add opportunities.

Clearly there are many applications of computer vision in real estate, but we still think the holy grail is using it to identify value-add deals in any market. Next we'll talk a bit about how HelloData is using computer vision today and how we plan to leverage it in the future.

How is HelloData.ai Using Computer Vision in Commercial Real Estate?

Of the above approaches, the most important for our algorithms are automating the analysis of property condition, extracting amenities and attributes, and detecting comparable properties with computer vision. Right now, our rent comp detection model is used by dozens of real estate investors, property managers and PropTech companies to deliver highly accurate rent comps in any U.S. market. We have several clients using our QualityScore API directly too, processing listing photos to assess condition and quality for market studies and appraisals.

In the near future, we're going to release a map-based feature that allows you to search for deals that are potentially undervalued based on their rents and the assessment or algorithms perform on their listing photos and street views. We've also discussed a "daily deals email" that looks at every new multifamily for sale listing that hits the market, analyzes its quality, condition and attributes, and determines the degree to which it is undervalued based on the market. Should be a gold mine for identifying value-add deals at scale. Stay tuned!

Check how to use HelloData's AI to find the best rent comps.

Marc Rutzen

Marc worked in real estate for 5 years before launching multifamily analytics startup Enodo, which he sold to Walker & Dunlop (NYSE: WD) in 2019. At W&D, he served as Chief Product Officer, developing products that helped source billions in loan volume. Outside of work, he enjoys reading, running, and spending time with family.

Recommended Articles