We are developing software products that combine CAE, AI, and the cloud. We also train and deliver AI models from customer data. In particular, the AI model creation from 3D CAD and CAE data is an unparalleled service.
Shape identification is the process of classifying the input 3D shapes into predefined classes, which are trained using supervised data.
Our 3D shape identification is not based on images, but on the 3D data itself.
Click here for a demo site of Shape identification.
Similarity analysis searches the database for shapes that are similar to the input shapes. It extracts features of the geometry and compares them with those of existing geometries in the database, using generative models such as VAE and GAN for training.
Our three-dimensional Similarity analysis is based on the three-dimensional data itself, not on images.
New shape synthesis
New shape synthesis automatically generates new shapes by blending feature vectors extracted from properly trained AI models such as used in similarity analysis, for multiple shapes. For example, for multiple existing car bodies, their tastes can be successfully blended to generate new car bodies.
Our three-dimensional new shape synthesis is determined from the three-dimensional data itself, not from the image.
Click here for a demo site of New shape synthesis.