Abstract

In recent years, there has been a tremendous success in transition of Machine Learning and Augmented Reality technologies from research institutions to industrial applications. These technologies are now fundamental to the industry as they make it possible to effectively solve a wide range of problems that were previously challenging to handle and enhance communication between engineers and computer systems. Current computer systems are capable of gaining full access to the data from various sensors deployed in industrial facilities. As a result, there is a growing interest in facilitating the retrieval of information about the components of process plants, which would help to make their operation more efficient. This work discusses an approach to perform 3D instance segmentation of industrial point clouds as well as its integration for the reconstruction of a digital representation of process plants. This study also explores several application scenarios which demonstrate the usefulness of the proposed approach.

Results/Implementation/Project Description

Conclusion

[ PDF (optional) ] 

[ Slides Kickoff/Final (optional)]