{"id":"arxiv:1801.02730","kind":"preprint","title":"Data Augmentation for Brain-Computer Interfaces: Analysis on Event-Related Potentials Data","authors":["Mario Michael Krell","Anett Seeland","Su Kyoung Kim"],"author_count":3,"year":2018,"published":"2018-01-09","venue":null,"cited_by":null,"doi":null,"arxiv_id":"1801.02730","nct_id":null,"source_url":"https://arxiv.org/abs/1801.02730","archive_url":"https://space-kitty.com/papers/arxiv:1801.02730","updated":"2018-01-09","pmid":null,"pmcid":null,"openalex_id":null,"pdf_url":"https://arxiv.org/pdf/1801.02730","abstract":"On image data, data augmentation is becoming less relevant due to the large amount of available training data and regularization techniques. Common approaches are moving windows (cropping), scaling, affine distortions, random noise, and elastic deformations. For electroencephalographic data, the lack of sufficient training data is still a major issue. We suggest and evaluate different approaches to generate augmented data using temporal and spatial/rotational distortions. Our results on the perception of rare stimuli (P300 data) and movement prediction (MRCP data) show that these approaches are feasible and can significantly increase the performance of signal processing chains for brain-computer interfaces by 1% to 6%.","abstract_license":"CC0-1.0 (arXiv descriptive metadata)","license":null,"is_open_access":false,"retracted":false,"peer_reviewed":false,"categories":["cs.CV","q-bio.NC"],"details":null,"provenance":"arxiv","attribution":"Abstract: arXiv descriptive metadata (CC0). Thank you to arXiv for use of its open access interoperability. This service was not reviewed or approved by, nor does it necessarily express or reflect the policies or opinions of, arXiv.","topics":[],"cited_in_dossiers":[]}