Launched in 2022 and recently updated in 2024 the ground-up reimagined PYP digital colorization tool focuses on automatic colorization of paintable surfaces within an user-uploaded photo.
Originally developed in 2008 and released to the public in 2009 Paint Your Place was an innovative digital tool empowering the user to be able visualize the Behr color library virtually in their home. Although ground breaking at the time, advancements in digital visualization tools over the years has driven user demand and expectation with the brands they use.
Objective:
Develop BEHR’s Paint Your Place into a best-in-class color visualization tool while improving the overall customer experience by automating previously necessary user interactions.
Insights from the project:
During the discovery phase of the project our agency partner and internal team investigated and hypothesized on different computer vision technologies such as point cloud, depth mapping, lidar, and photogrammetry. After researching the strengths and weaknesses of each as they pertain to time and cost to implement we landed on a solution and developed a proof of concept. Leveraging what I had developed in our static masking/colorization the team incorporated what I call normalization and regraying of images in the colorization process which greatly retains the light effects and accuracy in low value colored visualizations. Eventually refining a tech stack that was both cost effective at scale as well as novel and therefor protected as intellectual property.
Patent #12008766
Image processing to detect edges, walls, and surfaces for a virtual painting application
Advanced Computer Vision Recipe
Image Processing discipline that incorporates mathematics methods to find edges in a Digital Image and recolor segmented objects.
Depth Cloud Mapping
Each pixel relates to a distance between the image plane and the corresponding object in the RGB image.
Machine Learning Object Segmentation
A collection of related computer vision tasks that involve identifying objects in digital photographs.
OpenCV Canny Edge Detection
Algorithm that reduces images down series of thin lines around the edges of objects.
Linked Color Library Visualization
Linked to the API Color Library the RGB value is flooded into the object plane.
Light/Shadow and “Regraying”
Part of recipe that is evaluating light and shadow in addition to layering value across the entire image.
Aggregate Path
What appears to be simple is boiled down data into a simple approach and a clear path forward on what the digital tool should achieve for the user.
Audience
Xennial - Millennial DIY - DIFM users
Bought a new house | Remodeling | Room refresh | Just curious
Mapping out the User Experience
Turning 7 simple gray boxes into a digital tool. Mapping out user experience, functionality, user experience and aesthetic design. This process is iterative and collaborative as development hurdles are realized and functionality limitations are addressed.
Fitting into Behr.com
Integration inside of the website. How will the customer engage?