The power of V-Ray — in combination with the Click-Ins AI platform — is revolutionizing the insurance industry to help eliminate fraud. Discover how.
We love to see how our customers around the world are integrating V-Ray into their own software to utilize the full power of Chaos Group’s ray-traced rendering in their applications. With V-Ray's developer tools and flexible licensing, companies can bring their own applications to market and benefit from the high-quality renders that are a standard in architecture, commercials and movies.
Click-Ins is one of our remarkable clients embracing artificial intelligence and deep learning to tackle a global phenomenon: Auto insurance fraud. CTO and Co-Founder Dmitry Geyzersky set up the company with CEO Eugene Greenberg to find a tech-powered solution to a problem which is part of America’s $80 billion-plus annual insurance fraud bill — and which drives up premiums for customers.
The supersmart Click-Ins platform uses AI and machine learning to analyze digital photos of vehicular damage and generates a unique fingerprint for each particular area of destruction. This fingerprint is uploaded to the Click-Ins database where it can be matched to duplicate claims for the same damage, and assessed to find out the value of the claim. V-Ray played a crucial role in training the system by rapidly producing tons of photorealistic CG images of crash damage.
We talk to Dmitry Geyzersky, CTO and Co-Founder of Click-Ins about the technology that’s helping insurance companies and rental companies around the world tackle a serious problem.
Say hello to the insurance industry’s new heroes
Can you tell us a little about the technology?
DG: Click-Ins’ revolutionary, multidisciplinary technology is based on a combination of AI, deep learning, 3D modeling, applied mathematics and computer vision, enabling accurate recognition and analysis of every vehicle part, identifying all damage to the vehicle.
We don’t only localize any damage to the vehicle; we can measure it precisely. To our knowledge, we are the only company that can do this with a single photo without the use of special equipment or manual camera calibration.
AI plays an important role in the process by being able to spot patterns that can hardly be captured by a human eye — and do it at scale. Moreover, Click-Ins is fully committed to an ethical AI initiative by using mostly synthetic data for training its deep learning models.
The technology is being continuously tested during undergoing Proof-of-concept projects with multiple customers. The system is already integrated with the local insurance company Hachshara Insurance – the first production customer.
V-Ray technology helps us render an unprecedented level of realism in the synthetic data we use to train our deep learning models.
Dmitry Geyzersky, Co-Founder, Click-Ins
What is V-Ray’s role in the platform development and why did you decide to integrate our ray-tracing software for your renders?
DG: V-Ray technology helps us render an unprecedented level of realism in the synthetic data we use to train our deep learning models. Click-Ins plays an integral part in an ethical AI initiative. We believe companies using AI must play fair; therefore, we pay special attention to the data we are using for the training.
We never train our system on customers’ data. Here, V-Ray technology plays an important role in helping us achieve the highest level of realism and make our synthetic data look like real-world imagery, reducing the need for human labeling. V-Ray is very flexible and can be integrated into most complicated environments, such as the one developed by Click-Ins.
The integration of the V-Ray rendering engine into our system has brought a significant boost to the quality of our data.
You also use our library with pre-scanned VRscans materials — how did this improve your workflow?
DG: We use VRscan materials that were developed for the automotive industry to achieve an unprecedented level of realism.
Have you seen an improvement in the AI/software performance since you integrated V-Ray?
DG: The accuracy of deep learning models is directly affected by the quality of the training and validation data. In this regard, V-Ray integration significantly contributes to the improvement of our deep learning results.