INDEC services // enhance your competitive edge

INDEC is collaborating with the University of Applied Sciences and Arts Dortmund on various artificial intelligence (AI) projects. One key partnership is the BMBF DiffPro-ML competition organized by the university, in which students leverage AI technology to solve practical problems.

In the realm of benchmarking and particularly for GOBENCH, AI offers multiple potential benefits that are being explored by the research team in Dortmund and the INDEC development team.

The collaboration aims to test and validate the practicality of AI applications. For instance, AI can automate tasks, such as cropping images for the most visually appealing selection. With regard to the teardown process, AI can automatically create 3D models of objects by photographing them on a turntable. AI can also identify similar objects in a database, which simplifies part classification. Large Language Models (LLMs) can be employed to detect specific features, like counting LEDs on a circuit board or identifying components, thereby assisting users in data acquisition.

Furthermore, AI-powered chatbots can further enhance human interaction with the system.

Deep learning

• Self-supervised learning
• Computer vision (Vision transformer models)
• Large language models (such as Meta Llama3)
• Open-set deep-learning-based segmentation (SAM - Segment Anything) and object detection (YOLO-World)
• Neural radiance fields (NeRF) for view synthesis
• Multimodal embeddings (multimodal feature vectors) and vector databases
• At INDEC, we believe that AI is not a replacement for human expertise, but it can significantly enhance user effectiveness and efficiency in benchmarking.