SORDI Hackathon: idealworks Organizes Global AI Challenge

October 2022

In collaboration with well-known national and international players from the automotive and technology sectors, idealworks is hosting a global AI Challenge to train artificial intelligence in production based on synthetic data from the BMW iFactory. The open-source dataset SORDI (Synthetic Object Recognition Dataset for Industries) serves as the basis for the hackathon, which will start in November.

Together with its long-term partner NVIDIA Corporation, Microsoft Corporation, and the BMW Group’s BMW TechOffice MUNICH, idealworks will host the SORDI Hackathon 2022. More than 5,000 teams worldwide are called upon to train AI models to recognize objects from the BMW iFactory and solve real-world use cases from industry. For this purpose, the BMW Group is providing 200,000 synthetic images and videos from the BMW iFactory, thereby promoting the use and democratization of AI.

The registration period for the online hackathon has started on Monday, October 24, 2022. During two four-week phases, teams have the chance to submit their solutions and win prize money worth a total of 25,000 euros. The first phase, which begins on November 14, is dedicated to classic object recognition challenges in industry. With the help of a specially developed AI, the participants must be able to recognize almost 20 different objects as efficiently as possible. For this purpose, the BMW Group offers the use of the no-code AI pipeline developed by the BMW TechOffice MUNICH as well as the labeling tool Lite. In stage two, the ten best teams will compete against one another for an individual solution for AI-based analysis of logistics areas and the containers present there. Various experts from the four initiators will provide support for the contestants.

The BMW Group has already been deploying AI in various applications for quality assurance since 2019. The synthetic dataset SORDI allows much faster training of AI models and thus serves to significantly increase the cost efficiency of AI in production. The BMW TechOffice MUNICH’s rendering pipeline ensures the automatic creation of any number of synthesized images in photo-realistic HD quality including labels, which in turn allows for the realization of AI models with a high degree of robustness. In addition, integrated digital labels enable fundamental machine vision tasks for relevant areas of production, such as classification, object recognition, or segmentation.

Marc Kamradt, Head of BMW TechOffice MUNICH, explains: “We are effectively using our SORDI dataset to train AI in the production environment. With SORDI, we are able to further increase the efficiency of artificial intelligence and train AI models without any manual effort. Through the AI Challenge we share this approach with the GitHub community and hope for new insights from the participants’ solutions. The focus is on democratizing AI, which is why we will share all the results of the hackathon with the community afterwards.”

“SORDI revolutionizes the use of AI in industrial environments and significantly simplifies the development of intelligent, autonomous systems. The tasks to be solved in the hackathon correspond to the challenges we encounter in our daily work,” adds idealworks’ CTO Jimmy Nassif. “We are very excited to see the participants’ solutions.”

The SORDI dataset consists of more than 800,000 photorealistic images of production resources in 80 classes and includes objects of particular relevance in the core technologies of automotive engineering and logistics. Specifically designed for industrial environments, such as manufacturing and production, it significantly accelerates neural network training for device recognition in the factory environment and makes them even more efficient. More information about the hackathon is available at sordi.ai/hackathon.