The CDRIN team generates and integrates synthetic data, an artificial alternative to real-world data, to train various models for AI and computer vision. The goal is to explore the use of synthetic data in training large neural networks through innovative applications of game engines such as Unity and Unreal, alongside 3D rendering platforms such as Blender and Omniverse. Our expertise provides practical solutions to enhance these neural networks by avoiding the challenges of sourcing, acquiring, and labeling real data, particularly medical data, and by enabling scenarios that are impossible to capture in reality.
01
02
03
04
05
06
Computer vision, in its broadest sense, allows us to perceive, analyze, recognize, and process components of visual media. Currently, computer vision is primarily achieved using AI techniques (deep learning). It is natural for the graphics and AI teams at CDRIN to collaborate and leverage synthetic data generated in game engines to train computer vision models. These models assist our partners in various industries in their innovation objectives.
With our expertise in computer vision and AI, we can help companies create new tools and products, thereby contributing to their success.