For the research in the field of camera based vital parameter detection several datasets were created to advance the current state of the art.

We designed and constructed a measuring setup for the acquisition of lossless compressed video data. The video data, from up to 5 different cameras, can be synchronized with a powerful state of the art multimodal bio signal monitor device, capable of measuring up to 32 physiological signals simultaneously like the EEG, EMG, ECG, EOG, heart rate, relative blood flow, skin conductance, respiration, and temperature.The setup is equipped with low noise RGB, NIR cameras and two multispectral cameras, which can record up to 18 spectral band simultaneously.

Using this setup we generated robust data in multiple studies as a benchmark for several publications in the field of heart and respiration rate detection.

Publikationen

Michal Rapczynski, Philipp Werner, Ayoub Al-Hamadi, Continuous Low Latency Heart Rate Estimation from Painful Faces in Real Time, 23th International Conference on Pattern Recognition (2016)

Michal Rapczynski, Philipp Werner, Ayoub Al-Hamadi,  „Kontaktfreie kamerabasierte Messung der Herzrate in Echtzeit“, innteract conference (2016)

Michal Rapczynski, Frerk Saxen, Philipp Werner, Ayoub Al-Hamadi, „Der Einfluss von Hautfarbensegmentierung auf die kontaktfreie Schätzung von Vitalparameter“, 22. Workshop Farbbildverarbeitung- Ilmenau 2016

Rapczynski, M., Werner, P., Saxen, F., & Al-Hamadi, A. (2018, October). How the Region of Interest Impacts Contact Free Heart Rate Estimation Algorithms. In 2018 25th IEEE International Conference on Image Processing (ICIP) (pp. 2027-2031). IEEE.

Michal Rapczynski, Philipp Werner, Ayoub Al-Hamadi, Effects of Video Encoding on Camera Based Heart Rate Estimation, March 2019, IEEE Transactions on Biomedical Engineering PP(99):1-1 DOI: 10.1109/TBME.2019.2904326 

M. Fiedler, M. Rapczyński and A. Al-Hamadi, "Fusion-Based Approach for Respiratory Rate Recognition From Facial Video Images," in IEEE Access, vol. 8, pp. 130036-130047, 2020, doi: 10.1109/ACCESS.2020.3008687.