Publications by authors named "Lokendra Birla"
Despite the enormous achievements of Deep Learning (DL) based models, their non-transparent nature led to restricted applicability and distrusted predictions. Such predictions emerge from erroneous In-Distribution (ID) and Out-Of-Distribution (OOD) samples, which results in disastrous effects in the medical domain, specifically in Medical Image Segmentation (MIS). To mitigate such effects, several existing works accomplish OOD sample detection; however, the trustworthiness issues from ID samples still require thorough investigation.
View Article and Find Full Text PDFAND-rPPG: A novel denoising-rPPG network for improving remote heart rate estimation.
Comput Biol Med
February 2022
Article Synopsis
- Heart rate (HR) estimation using Remote Photoplethysmography (r-PPG) has become crucial, especially for telehealth during the COVID-19 pandemic, allowing heart rate measurement from non-contact face videos.
- Current r-PPG methods struggle with facial deformations from expressions, leading to inaccurate HR estimates because they treat all facial areas the same and ignore how expressions create different types of noise.
- The paper presents a new method called AND-rPPG, which uses Action Units (AUs) to effectively denoise the signals based on facial expressions, resulting in better HR estimation than existing techniques and enhancing their performance when integrated.