This paper presents a health monitoring method by estimating vital signs using an RGB camera. The rPPG signal is used to estimate the physical parameters with the help of a non-invasive smartphone camera. The vital signs of human are very important especially in health monitoring applications. In this paper, the deep learning-based algorithm has been used to estimate the vital signs using rPPG signal based on RGB frames camera video. The convolutional neural network (CNN) has been used to estimate the vital sign such as heart rate and breathing rate, their ratio, and Sp02. The features were extracted from the last convolutional layer (C5) of the pre-trained VGG16 model. The average of the blood intensity variation extracted as a feature matrix from the last convolutional layer represents the rPPG signal which further used to estimate the vital signs. The results show that proposed technique produced better performance as compared to existing standard and conventional vital signs estimation techniques. © 2019 IEEE.