This is the website of the ICPR 2016 paper: S. Bobbia, Y. Benezeth, J. Dubois, "Remote Photoplethysmography Based on Implicit Living Skin Tissue Segmentation, "IEEE International Conference on Pattern Recognition (ICPR), 2016. (pdf)
Region of interest selection is an essential part for remote photoplethysmography (rPPG) algorithms. Most of the time, face detection provided by a supervised learning of physical appearance features coupled with skin detection is used for region of interest selection. However, both methods have several limitations and we propose to implicitly select living skin tissue via their particular pulsatility feature.
The input video stream is decomposed into several temporal superpixels from which pulse signals are extracted. Pulsatility measure for each temporal superpixel is then used to merge pulse traces and estimate the photoplethysmogram signal. This allows to select skin tissue and furthermore to favor areas where the pulse trace is more predominant.
Figure below illustrates pulsatility measures estimated from various temporal superpixels (blue means low pulsatility measures and yellow/orange is high)
Our method is a 4-step pipeline which are:
(1) Input video stream is (2) decomposed into temporally consistent superpixels.
(3) Tentative rPPG signal is extracted from each temporal superpixel
(4) A pulsatility measure is estimated for each ROI
(5) A weighted average of all the tentative rPPG signals is finally computed.
Link to download the complete video dataset used in the paper is available on request. A basic Matlab implementation can also be provided to read ground truth data acquired with a pulse oximeter.
Please go to the UBFC-RPPG dataset webpage (we have used dataset_1 in this paper).
