Research Spotlight: Mining Video Data

We are always excited to learn about new research from our academic partners. Today, we would like to highlight an interesting paper about the design and implementation of video-based process mining that Wolfgang Kratsch, Fabian König, and Maximilian Röglinger have published in the Decision Support Systems journal.

Manual activities often leave blind spots in the process analysis because you cannot see them in the data. Especially for manufacturing processes, there is an opportunity to use video data to fill these blind spots. The video data contains valuable process-related information, but there is no standardized approach to creating event logs from unstructured video data yet.

The authors developed a reference architecture that bridges the gap between computer vision and process mining. The computer vision capabilities that are the basis for extracting the process-related information from video data are shown in Figure 1 below (click on the image to see a larger version). The extracted low-level events are then aggregated to the level of detail required for the process mining use case.

Figure 1. Illustration of computer vision capabilities

You can freely access their paper until 17 June 2022 here. Thanks to Wolfgang, Fabian, and Maximilian for allowing us to share their research!

Anne Rozinat

Anne Rozinat

Market, customers, and everything else

Anne knows how to mine a process like no other. She has conducted a large number of process mining projects with companies such as Philips Healthcare, Océ, ASML, Philips Consumer Lifestyle, and many others.