Auflistung nach Autor:in "Atzmueller, Martin"
1 - 5 von 5
Treffer pro Seite
Sortieroptionen
- KonferenzbeitragActive-learning-driven deep interactive segmentation for cost-effective labeling of crop-weed image data(43. GIL-Jahrestagung, Resiliente Agri-Food-Systeme, 2023) Sikouonmeu, Freddy; Atzmueller, MartinActive learning has shown its reliability in (semi-)supervised machine learning tasks to reduce the labeling cost for large datasets in various areas. However, in the agricultural field, despite past attempts to reduce the labeling cost and the burden on the labeler in acquiring image labels, the load during the acquisition of pixel-level labels for semantic image segmentation tasks remains high. Typically, the respective pixel-level masks are acquired manually by drawing polygons over irregular and complex-shaped object boundaries. In contrast, this paper proposes a method leveraging the power of a click-based deep interactive segmentation model (DISEG) in an active learning approach to harvest high-quality image segmentation labels at a low cost for training a real-time task model by only clicking on the objects’ fore- and background surfaces. Our first experimental results indicate that with an average of 3 clicks per image object and using only 3% of the unlabeled dataset, we can acquire pixel-level labels with good quality at low cost.
- KonferenzbeitragBehavioral Analysis on Socio-Spatial Interaction Networks concerning User Preferences, Interactions and their Perception(Proceedings of the 23rd International Workshop on Personalization and Recommendation on the Web and Beyond, 2019) Atzmueller, Martin; Güven, Cicek; Masiala, Spyroula; Mackenbach, Rick; Shayan, Parisa; Liebregts, WernerThis paper investigates socio-spatial interaction networks for user modeling: We analyze preferences and perceptions of socio-proximity human interactions in relation to the observed interactions. The analysis is performed on a real-world dataset capturing interaction networks using wearable sensors coupled with self-report questionnaires about preferences and perception of those interactions.
- KonferenzbeitragDescriptive Network Modeling and Analysis for Investigating User Acceptance in a Learning Management System Context(Proceedings of the 23rd International Workshop on Personalization and Recommendation on the Web and Beyond, 2019) Shayan, Parisa; Rondinelli, Roberto; van Zaanen, Menno; Atzmueller, MartinLearning Management Systems (LMSs) play a significant role in educational technology. In this paper, we analyze different approaches in order to investigate the acceptance of an LMS. Utilizing questionnaire information structured on the Technology Acceptance Model (TAM), we apply descriptive network modeling and analysis complementing basic statistical analysis in order to identify specific patterns in the user data. We present the applied analysis methodology in detail, and demonstrate the connection to user modeling:here, descriptive statistics indicate student satisfaction with the usage (acceptance level) as a whole; network analysis indicates the level of variability w.r.t. the user questions, while specific patterns or motifs show the satisfaction levels for the different networks.
- KonferenzbeitragHow to select a suitable tool for a software development project: three case studies and the lessons learned(Software Engineering 2013 - Workshopband, 2013) Kibanov, Mark; Erdmann, Dominik J.; Atzmueller, MartinThis paper describes a framework for evaluating and selecting suitable software tools for a software project, which is easily extendable depending on needs of the project. For an evaluation, we applied the presented framework in three different projects. These projects use different software development methods (from classical models to Scrum) in different environments (industry and academia). We discuss our experiences and the lessons learned.
- ZeitschriftenartikelMining Social Media(Informatik-Spektrum: Vol. 35, No. 2, 2012) Atzmueller, Martin