SynthAI Demo #1: Upload CAD Data & Start Training [VIDEO]

By Kelly Gallagher

In this first demo series of SynthAI, learn how to train object detection machine learning with synthetic data in three easy steps.

Synthetic data is transforming the machine learning world before our very eyes. This artificial computer-generated data can stand-in for data obtained from the real world and is used for a variety of vision-based automation use cases such as robotic bin picking, sorting, palletizing, quality inspection, and others. Although the usage of machine learning for vision-based automations continues to expand, many industries face the challenge with the successful implementation within their computer vision applications. That is because the time required to generate synthetic data from real-world data, such as collecting thousands of images, and the challenges associated with accurately annotating the different items within those images is cumbersome and costly. 

Factory engineers need a way to quickly ramp up their AI and machine learning algorithms so that simulations can be deployed in a much more efficient manner. Discover how Siemens SynthAI, AI-driven cloud-based platform, solves this complex problem by watching the video demo below. This web-based solution allows factory engineers, like you, to replicate a digital twin of the part or object being manufactured and accelerate machine learning algorithms to perform various tasks like detecting objects or defects more effectively. SynthAI generates synthetic data that can train machine learning models to detect a product in real life. This capability allows manufacturing engineers to quickly and easily use the trained model to detect the object within the image that is obtained from the physical shop floor. 

In this first tutorial of our SynthAI series, see how you can train object detection machine learning in three easy steps. In this training, we focus on the object detection task which identifies every single instance of an object within the image and automating notations so that the simulator is working off of pixel-perfect annotations. All you need to do is create a new project, upload a CAD file that represents the object that you would like to train the model on, then start the desired training session.  

When we combine synthetic training data for computer vision-based manufacturing solutions, we are able to make quality training data vastly more accessible and affordable, all while accelerating faster simulations and industrial deployments on the manufacturing floor. In our next demo, you will see how to view the training results, download the synthetic dataset and trained model. Stay tuned to discover how we’re making training data readily available for AI-powered computer vision systems. What was once seen as a major obstacle in the development of AI-powered industrial systems is now made possible with SynthAI. Interested in learning more about the demonstration? Visit our Tecnomatix Community for more information for a step-by-step training of the tutorial.

Stay tuned to learn the latest about how SynthAI is transforming AI-powered systems for real world industrial use cases.

Siemens SynthAI_demo

Request Early Access to SynthAI

SynthAI is a cloud solution that allows manufacturing engineers to easily generate synthetic data for training ML models for computer vision applications.

Use synthetic data to train object detection machine learning in three easy steps.

Leave a Reply

This article first appeared on the Siemens Digital Industries Software blog at