All Eyes on the Workflow: Automated and Efficient Event Discovery from Video Streams

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Pegoraro, Marco, Seng, Jonas, Heller, Dustin, van der Aalst, Wil M. P., Kersting, Kristian
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866910162771509248
author Pegoraro, Marco
Seng, Jonas
Heller, Dustin
van der Aalst, Wil M. P.
Kersting, Kristian
author_facet Pegoraro, Marco
Seng, Jonas
Heller, Dustin
van der Aalst, Wil M. P.
Kersting, Kristian
contents Disciplines such as business process management and process mining aid organizations by discovering insights about processes on the basis of recorded event data. However, an obstacle to process analysis is data multi-modality: for instance, data in video form are not directly interpretable as events. In this work, we present SnapLog, an approach to extract event data from videos by converting frames to feature vectors using image embeddings and performing temporal segmentation through frame-wise similarity matrices. A generalized few-shot classification is then used to assign labels to the video segments, yielding labeled, timestamped sub-sequences of frames that are interpretable as events. Conventional process mining techniques can be used to analyze the resulting data. We show that our approach produces logs that accurately reflect the process in the videos.
format Preprint
id arxiv_https___arxiv_org_abs_2604_22476
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle All Eyes on the Workflow: Automated and Efficient Event Discovery from Video Streams
Pegoraro, Marco
Seng, Jonas
Heller, Dustin
van der Aalst, Wil M. P.
Kersting, Kristian
Computer Vision and Pattern Recognition
Machine Learning
Disciplines such as business process management and process mining aid organizations by discovering insights about processes on the basis of recorded event data. However, an obstacle to process analysis is data multi-modality: for instance, data in video form are not directly interpretable as events. In this work, we present SnapLog, an approach to extract event data from videos by converting frames to feature vectors using image embeddings and performing temporal segmentation through frame-wise similarity matrices. A generalized few-shot classification is then used to assign labels to the video segments, yielding labeled, timestamped sub-sequences of frames that are interpretable as events. Conventional process mining techniques can be used to analyze the resulting data. We show that our approach produces logs that accurately reflect the process in the videos.
title All Eyes on the Workflow: Automated and Efficient Event Discovery from Video Streams
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2604.22476