모바일넷V3를 이용한 어텐션 기반의 심박수 예측
Attention-Based Heart Rate Estimation using MobilenetV3
3
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
딥러닝의 발전은 의료 분야에서도 다양한 응용을 가능하게 하고 있으며 이러한 애플리케이션중에 심박수 측정은 개인의 건강을 관리하기 위한 필수적인 아이템이라 할 수 있다. 광혈류 측정을 이용한 기존 방법의 경우 스마트워치 같은 장비의 착용이 필수적이다. 그러나 최근 딥러닝 기술의 발전은 비침습식으로 원격에서 사용자의 얼굴 이미지를 분석하여 심박수를 높은 성능으로측정가능하게 한다. 본 연구에서는 모바일 환경에서 사용 가능한 경량화된 심박수 추정 방법론을제안한다. 이 방법론은 2D 컨볼루션에 기반한 특화된 2채널 네트워크 구조를 사용하여, 혈류와 근육 수축으로 인한 얼굴의 미세한 움직임과 색상 변화를 고려한다. 제안하는 네트워크 구조는 이미지 특성을 분석하는 인코더와 혈류량 파동을 예측하는 회귀 레이어로 구성되어있다. 이러한 복합적인 특성을 동시에 분석함으로써, 제한된 컴퓨팅 리소스를 가진 환경에서도 심박수를 정확하게 추정할 수 있다. 이 연구의 접근 방식은 침습적인 기술 없이도 심박수를 효과적으로 모니터링할 수 있는 새로운 경로를 제공할 것으로 예상한다.
Abstract
The advent of deep learning technologies has led to the development of various medical applications, making healthcare services more convenient and effective. Among these applications, heart rate estimation is considered a vital method for assessing an individual's health. Traditional methods, such as photoplethysmography through smart watches, have been widely used but are invasive and require additional hardware. Recent advancements allow for contactless heart rate estimation through facial image analysis, providing a more hygienic and convenient approach. In this paper, we propose a lightweight methodology capable of accurately estimating heart rate in mobile environments, using a specialized 2-channel network structure based on 2D convolution. Our method considers both subtle facial movements and color changes resulting from blood flow and muscle contractions. The approach comprises two major components: an Encoder for analyzing image features and a regression layer for evaluating Blood Volume Pulse. By incorporating both features simultaneously our methodology delivers more accurate results even in computing environments with limited resources. The proposed approach is expected to offer a more efficient way to monitor heart rate without invasive technology, particularly well-suited for mobile devices.
참고문헌 (16)
- Liu, Xin, "Multi-task temporal shift attention networks for on-device contactless vitals measurement", Advances in Neural Information Processing Systems, 33, pp.19400~19411 (2020)
- Liu, Xin, "MetaPhys: few-shot adaptation for non-contact physiological measurement", Proceedings of the conference on health, inference, and learning (2021)
- Yu, Zitong, "Transrppg: Remote photoplethysmography transformer for 3d mask face presentation attack detection", IEEE Signal Processing Letters, 28, pp.1290~1294 (2021)
- Lee, Eugene, "Meta-rppg: Remote heart rate estimation using a transductive meta-learner", Computer Vision–ECCV 2020: 16th European Conference, pp.2020~ (2020), Springer International Publishing
- Andrew Howard, "Searching for mobilenetv3", Proceedings of the IEEE, 2019, ~
- Mark Sandler, "Mobilenetv2: Inverted residuals and linear bottlenecks", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.4510~4520 (2018)
- Ningning Ma, "Shufflenet v2: Practical guidelines for efficient cnn architecture design" (2018), The European Conference on Computer Vision (ECCV)
- Xiangyu Zhang, "Shufflenet: An extremely efficient convolutional neural network for mobile devices", The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2018)
- Yunsheng Li, "Micronet: Improving image recognition with extremely low flops", International Conference on Computer Vision (2021)
- Chen, Yinpeng, "Mobile-former: Bridging mobilenet and transformer", pp.2022, Proceedings of the IEEE
- Li, Yanyu, "Efficientformer: Vision transformers at mobilenet speed", Advances in Neural Information Processing Systems, 35, pp.12934~12949 (2022)
- D. Sinha, "Thin MobileNet : An Enhanced MobileNet Architecture", 2019 IEEE 10th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference(UEMCON), New York, NY, USA, pp.0280~0285 (2019)
- Sabour, Rita Meziati, "Ubfc-phys: A multimodal database for psychophysiological studies of social stress", IEEE Transactions on Affective Computing (2021)
- Tsou, Yun-Yun, "Siamese-rPPG network : Remote photoplethysmography signal estimation from face videos", Proceedings of the 35th annual ACM symposium on applied computing (2020)
- Stricker, "Non-contact video-based pulse rate measurement on a mobile service robot", The 23rd IEEE International Symposium on Robot and Human Interactive Communication (2014), IEEE