Print Email Facebook Twitter Improving state estimation through projection post-processing for activity recognition with application to football Title Improving state estimation through projection post-processing for activity recognition with application to football Author Ciszewski, M.G. (TU Delft Statistics) Söhl, J. (TU Delft Statistics) Jongbloed, G. (TU Delft Statistics) Date 2023 Abstract The past decade has seen an increased interest in human activity recognition based on sensor data. Most often, the sensor data come unannotated, creating the need for fast labelling methods. For assessing the quality of the labelling, an appropriate performance measure has to be chosen. Our main contribution is a novel post-processing method for activity recognition. It improves the accuracy of the classification methods by correcting for unrealistic short activities in the estimate. We also propose a new performance measure, the Locally Time-Shifted Measure (LTS measure), which addresses uncertainty in the times of state changes. The effectiveness of the post-processing method is evaluated, using the novel LTS measure, on the basis of a simulated dataset and a real application on sensor data from football. The simulation study is also used to discuss the choice of the parameters of the post-processing method and the LTS measure. Subject Activity recognitionPerformance measuresPost-processingWearable sensors To reference this document use: http://resolver.tudelft.nl/uuid:0ed0d0ed-19d1-4a95-b9c7-cd1444745130 DOI https://doi.org/10.1007/s10260-023-00696-z ISSN 1618-2510 Source Statistical Methods and Applications, 32 (5), 1509-1538 Part of collection Institutional Repository Document type journal article Rights © 2023 M.G. Ciszewski, J. Söhl, G. Jongbloed Files PDF s10260_023_00696_z.pdf 2.47 MB Close viewer /islandora/object/uuid:0ed0d0ed-19d1-4a95-b9c7-cd1444745130/datastream/OBJ/view