This work addresses the problem of recovering the morphology of blood volume pulse (BVP) information from a video of a person's face. Video-based remote plethysmography methods have shown promising results in estimating vital signs such as heart rate and breathing rate. However, recovering the instantaneous pulse rate signals is still a challenge for the community. This is due to the fact that most of the previous methods concentrate on capturing the temporal average of the cardiovascular signals. In contrast, we present an approach in which BVP signals are extracted with a focus on the recovery of the signal morphology as a generalized form for the computation of physiological metrics. We also place emphasis on allowing natural movements by the subject. Furthermore, our system is capable of extracting individual BVP instances with sufficient signal detail to facilitate candidate re-identification. These improvements have resulted in part from the incorporation of a robust skin-detection module into the overall imaging-based photoplethysmography (iPPG) framework. We present extensive experimental results using the challenging UBFC-Phys dataset and the well-known COHFACE dataset. The source code is available at https://github.com/yogeshd21/CVPM-2023-iPPG-Paper. / Master of Science / In this work we are trying to study and recover human health related metrics and the physiological signals which are at the core for the derivation of such metrics. A well know form of physiological signals is ECG (Electrocardiogram) signals and for our research we work with BVP (Blood Volume Pulse) signals. With this work we are proposing a Deep Learning based model for non-invasive retrieval of human physiological signals from human face videos. Most of the state of the art models as well as researchers try to recover averaged cardiac pulse based metrics like heart rate, breathing rate, etc. without focusing on the details of the recovered physiological signal. Physiological signals like BVP have details like systolic peak, diastolic peak and dicrotic notch, and these signals also have applications in various domains like human mental health study, emotional stimuli study, etc. Hence with this work we focus on retrieval of the morphology of such physiological signals and present a quantitative as well as qualitative results for the same. An efficient attention based deep learning model is presented and scope of reidentification using the retrieved signals is also explored. Along with significant implementations like skin detection model our proposed architecture also shows better performance than state of the art models for two very challenging datasets UBFC-Phys as well as COHFACE. The source code is available at https://github.com/yogeshd21/CVPM-2023-iPPG-Paper.

No description available.

This study was aimed at determining whether the digital volume pulse waveform using the Pulse Sensor can be used to extract features related to arterial compliance. The Pulse Sensor, a low-cost photoplethysmograph, measures green light reflection in the finger and generates output, which is indicative of blood flow and can be read by the low-cost Arduino UNO™. The Pulse Sensor code was modified to increase the sampling frequency and to capture the data in a file, which is subsequently used for waveform analysis using programs written in the R system. Waveforms were obtained using the Pulse Sensor during two 30-s periods of seated rest, in each of 44 participants, who were between the ages of 20 and 80 years. For each cardiac cycle, the first four derivatives of the waveform were calculated and low-pass filtered by convolution before every differentiation step. The program was written to extract 19 features from the pulse waveform and its derivatives. These features were selected from those that have been reported to relate to the physiopathology of hemodynamics. Results indicate that subtle features of the pulse waveform can be calculated from the fourth derivative. Feature misidentification occurred in cases of saturation or low voltage and resulted in outliers; therefore, trimmed means of the features were calculated by automatically discarding the outliers. There was a high efficiency of extraction for most features. Significant relationships were found between several of the features and age, and systolic, diastolic, and mean arterial blood pressure, suggesting that these features might be employed to predict arterial compliance. Further improvements in experimental design could lead to a more detailed evaluation of the Pulse Sensor with respect to its capability to predict factors related to arterial compliance.