Computer Software
Auto-insurance ISV
Based in California, our client is an automotive industry software provider. The company’s focus was to make insurance claim management and collision repair more straightforward, efficient, and accessible—through smart technology solutions augmented by our industry expertise.
Our team utilized an ODC architecture to effectively detect and classify objects within images. For the AI & ML technology solution, we used our image repository to train the model with sufficient data. Once the model completed the training, we customized it for damage detection. It enabled the model to accurately identify and classify any damage in the images.
Using custom neural network architecture to classify images
For every damaged part visible in the image
The highest per-panel average for the ‘Replace’ operation at 0.94.
Our team utilized an ODC architecture to effectively detect and classify objects within the images. For the AI and ML technology solution, we used our image repository to train the model with a sufficient amount of data. Once the training was completed, we customized it specifically for damage detection. It enabled the model to accurately identify and classify any damage present in the images.
Powered by a custom neural network architecture, we worked with our client to build an AI-powered application that quickly and accurately identifies vehicle damage from an image and immediately recommends the most appropriate action.
Our pathbreaking solution was the collision repair industry’s first extended reality (XR) solution designed to digitize and accelerate manual activities in the repair process. We built the IV wearable device on the RealWear HMT -1 platform. The high-resolution camera enabled repair professionals to capture photos and videos quickly. It also automatically extracted text-based information, saving time and preventing errors. An integrated accelerometer tracked the head movements to facilitate navigating detailed repair procedures, creating a hands-free experience optimized to support proper, safe repair.
We deployed our highly scalable solution on AWS Lambda. Our client could therefore handle up to 40 claims per minute using advanced image preprocessing— with early adopters testing the machine learning recommendations. Our collaboration also won Silver in three major categories of international design awards recently.
From a modest project in 2017 to an extensive and comprehensive development and support program today, ValueLabs has surpassed the status of being merely an IT partner. We have progressed to a strategic OneCompany® relationship with the client and have successfully introduced efficiency-improving artificial intelligence and machine learning solutions. Intending to meet growing consumer demands, the client continue to use our deep technical expertise to push the limits of automotive claim management innovation.
Accuracy in panel-level damage detection
Less time spent processing claims
Claims per minute on average as end-to-end solution deployed on AWS with robust inference pipelines
Learn from our experience and expertise in AI & Machine Learning
01
Our sales managers reach out to you.
02
Our experts get in touch with you to understand your requirements.
03
We discuss and propose project estimates and timelines.
5 Min read
