TR2023-116

Unrolled IPPG: Video Heart Rate Esitmation via Unrolling Proximal Gradient Descent


    •  Shenoy, V., Marks, T.K., Mansour, H., Lohit, S., "Unrolled IPPG: Video Heart Rate Esitmation via Unrolling Proximal Gradient Descent", IEEE International Conference on Image Processing (ICIP), September 2023.
      BibTeX TR2023-116 PDF Video
      • @inproceedings{Shenoy2023sep,
      • author = {Shenoy, Vineet and Marks, Tim K. and Mansour, Hassan and Lohit, Suhas},
      • title = {Unrolled IPPG: Video Heart Rate Esitmation via Unrolling Proximal Gradient Descent},
      • booktitle = {IEEE International Conference on Image Processing (ICIP)},
      • year = 2023,
      • month = sep,
      • url = {https://www.merl.com/publications/TR2023-116}
      • }
  • MERL Contacts:
  • Research Areas:

    Computer Vision, Machine Learning

Abstract:

Imaging photoplethysmography (iPPG) is the process of estimating a person’s heart rate from video. In this work, we propose Un- rolled iPPG, in which we integrate iterative optimization updates with deep learning-based signal priors to estimate the pulse wave- form and heart rate from facial videos. We model the signal extracted from video as the sum of an underlying pulse signal and noise, but instead of explicitly imposing a handcrafted prior (e.g., sparsity in the frequency domain) on the signal, we learn priors on the signal and noise using neural networks. We solve for the underlying pulse sig- nal by unrolling proximal gradient descent; the algorithm alternates between gradient descent steps and application of learned denoisers, which replace handcrafted priors and their proximal operators. Using this method, we achieve state-of-the-art heart rate estimation on the challenging MMSE-HR dataset.

 

  • Related Videos