Search

Video-based Photoplethysmography (VPPG) offers the capability to measure heart rate (HR) from facial videos. However, the reliability of the HR values extracted through this method remains uncertain, especially when videos are affected by various disturbances. Confronted by this challenge, we introduce an innovative framework for VPPG-based HR measurements, with a focus on capturing diverse sources of uncertainty in the predicted HR values. In this context, a neural network named HRUNet is structured for HR extraction from input facial videos. Departing from the conventional training approach of learning specific weight (and bias) values, we leverage the Bayesian posterior estimation to derive weight distributions within HRUNet. These distributions allow for sampling to encode uncertainty stemming from HRUNet's limited performance. On this basis, we redefine HRUNet's output as a distribution of potential HR values, as opposed to the traditional emphasis on the single most probable HR value. The underlying goal is to discover the uncertainty arising from inherent noise in the input video. HRUNet is evaluated across 1,098 videos from seven datasets, spanning three scenarios: undisturbed, motion-disturbed, and light-disturbed. The ensuing test outcomes demonstrate that uncertainty in the HR measurements increases significantly in the scenarios marked by disturbances, compared to that in the undisturbed scenario. Moreover, HRUNet outperforms state-of-the-art methods in HR accuracy when excluding HR values with 0.4 uncertainty. This underscores that uncertainty emerges as an informative indicator of potentially erroneous HR measurements. With enhanced reliability affirmed, the VPPG technique holds the promise for applications in safety-critical domains.

Objective: Recent research indicates that facial epidermis color varies with the rhythm of heat beats. It can be captured by consumer-level cameras and, astonishingly, be adopted to estimate heart rate (HR). The HR estimated remains not as precise as required in a practical environment where illumination interference, facial expressions, or motion artifacts are involved, although numerous methods have been proposed in the last few years. A novel algorithm is proposed to make non-contact HR estimation technique more robust., Approach: First, the face of the subject is detected and tracked to follow the head movement. The facial region then falls into several blocks, and the chrominance feature of each block is extracted to establish a raw HR sub-signal. Self-adaptive signal separation is performed to separate the noiseless HR sub-signals from raw sub-signals. On that basis, the noiseless sub-signals full of HR information are selected using a weight-based scheme to establish the holistic HR signal, from which the average HR is computed adopting wavelet transform and data filtering., Main Results: Forty subjects took part in our experiments, whose facial videos were recorded by a normal webcam with the frame rate of 30 fps under ambient lighting conditions. The average HR estimated by our method correlates strongly with ground truth measurements, as indicated in experimental results measured in a static scenario with the Pearson correlation r  =  0.980 and a dynamic scenario with the Pearson correlation r  =  0.897. In addition, our method, compared to the newest method, decreases the error rate by 38.63% and increases the Pearson correlation by 15.59%., Significance: This work proposes a robust method for non-contact HR measurement in a realistic environment. Results of comparative experiments indicate that our method out-performs state-of-the-art non-contact HR estimation methods in realistic environments.

Recent studies have shown that changes in human facial skin color due to varying blood flows can be recorded on videos by consumer-level cameras and, interestingly, can be used to measure pulse rates in a noncontact manner. However, it remains unclear whether the video-based pulse rate detection is feasible when videos are recorded in unstable lighting conditions. To address unstable illumination, a novel algorithm named adaptive spatiotemporal homomorphic filtering (ASTHF) is proposed in this article, with a focus on the separation of pulse signals and illumination-induced noise. This method first employs the homomorphic filtering framework to make illumination-induced noise in a video linearly separable. Subsequently, it incorporates a spatiotemporal filter into the framework, aiming to remove illumination-induced noise from both the time domain and the space domain of the video. To adapt to diverse illumination in the real world, the filter parameters are determined adaptively according to the sparsity in frequency spectrum of pulse signals. ASTHF is tested on two data sets, including 406 video samples. The test results show that ASTHF has stronger robustness to unstable illumination than state-of-the-art methods, improving the Pearson’s correlation coefficient by 5% in a challenging scenario compared with the best contrasting method. The proposed method demonstrates promise to achieve noncontact pulse rate detection from facial videos involving unsteady illumination. [ABSTRACT FROM AUTHOR]

Objective: Modern consumer-level cameras can detect subtle changes in human facial skin color due to varying blood flow; they are beginning to be used as noncontact devices to detect pulse waves. Little, however, do we know about their capacity to perform pulse wave detection when the recorded faces are unstable. Methods: Here, we propose a novel method that can extract pulse waves from videos with drastic facial unsteadiness such as head twists and alternating expressions. The method first uses chrominance characteristics in multiple facial sub-regions to construct a raw pulse matrix. Subsequently, it employs a disturbance-adaptive orthogonal matching pursuit (DAOMP) algorithm to recover the underlying pulse matrix corrupted by facial unsteadiness. Results: To evaluate the efficacy of the method, we perform analyses on two datasets including 268 samples from 67 testing subjects. The results demonstrate that the proposed method outperforms state-of-the-art algorithms, especially in the terrain where drastic facial unsteadiness is present. Conclusion: The proposed framework shows promise to achieve videos-based noncontact pulse wave detection from both steady and unsteady faces recorded by consumer-level cameras. Significance: By employing the proposed method, disturbance robustness in noncontact pulse wave detection can be significantly improved. [ABSTRACT FROM AUTHOR]

The locations and breathing signal of people in disaster areas are significant information for search and rescue missions in prioritizing operations to save more lives. For detecting the living people who are lying on the ground and covered with dust, debris or ashes, a motion magnification-based method has recently been proposed. This current method estimates the locations and breathing signal of people from a drone video by assuming that only human breathing-related motions exist in the video. However, in natural disasters, background motions, such as swing trees and grass caused by wind, are mixed with human breathing, that distort this assumption, resulting in misleading or even no life signs locations. Therefore, the life signs in disaster areas are challenging to be detected due to the undesired background motions. Note that human breathing is a natural physiological phenomenon, and it is a periodic motion with a steady peak frequency; while background motion always involves complex space-time behaviors, their peak frequencies seem to be variable over time. Therefore, in this work we analyze and focus on the frequency properties of motions to model a frequency variability feature used for extracting only human breathing, while eliminating irrelevant background motions in the video, which would ease the challenge in detection and localization of life signs. The proposed method was validated with both drone and camera videos recorded in the wild. The average precision measures of our method for drone and camera videos were 0.94 and 0.92, which are higher than that of compared methods, demonstrating that our method is more robust and accurate to background motions. The implications and limitations regarding the frequency variability feature were discussed. [ABSTRACT FROM AUTHOR]