Background:
21Moves developed a Rubik’s Cube solver mobile app and was looking to augment their app features. They had a specific focus on refining the user experience.
Before:
- Their initial build of the scanner was sufficient to get them off the ground, however had very limited capabilities.
- The app struggled with accurately interpreting images affected by glare, leading to compromised results.
- The scanner encountered difficulty in accurately recognising certain cubes, particularly due to variations in the size and shape of their individual pieces.
- The app confused red and oranges on occasions, making it impossible to scan accurately.
- Images with deep shadows presented challenges in extracting meaningful information.
After:
- The scanner now accurately interprets various cube configurations, even in challenging conditions.
- Issues related to glare are effectively mitigated, ensuring precise results under all lighting conditions.
- Accurately distinguish between colours, eliminating previous confusion between red and orange.
- Deep shadows no longer hinder scanning accuracy.
- Increase of users using the Computer vision app from 60% to 80%, solidifying its positioning in the market.
- More than 500k downloads, over 25000 reviews and a 4.7 rating.
Process:
- Analyze the customer’s image processing pipeline to identify potential points of failure.
- Design a new image processing pipeline able to handle a wider range of situations making the APP easier to use and more reliable.
They helped improve core aspects of our computer vision pipeline. Not only did they improve key functionality like glare detection, they provided a great sounding board for determining overall strategy with the product moving forward. We were able to go from 60% of users successfully using our computer vision scanner, to 80%!
