Vital sign monitoring is essential to prevent diseases related to cardiac and respiratory activity. Remote photoplethysmography from facial video streams is a widely used non-contact monitoring approach, based on the detection of subtle changes in skin colour caused by cardiovascular activity. This paper proposes a photoplethysmography pipeline for Heart Rate, Breath Rate, and Blood Oxygen Saturation estimation integrated into a commercial and low cost mobile device, using its frontal camera for data acquisition. The smartphone's sensing, processing, and gateway capabilities enable its use anywhere and anytime, without requiring complex setups or dedicated specialists. The algorithmic pipeline includes two main blocks: pre-processing and feature extraction/vital signs estimation, integrating various algorithmic steps for photoplethysmographic signal extraction and the estimation of the previously introduced vital signs. The validation of the system was performed on 20 individuals, and the performance were calculated in terms of Mean Absolute Error and Root Mean Square Error, considering varying ambient illumination, distance and orientation of the user's face with respect to the mobile camera. The results demonstrate that the implemented pipeline provides accurate vital signs estimation comparable to wearable sensors, enabling a real-time and non-invasive monitoring.