Learning Motion-Robust Remote Photoplethysmography through Arbitrary Resolution Videos

Authors: Jianwei Li, Zitong Yu, Jingang Shi

AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on three benchmark datasets (UBFC-r PPG, COHFACE and PURE) demonstrate the superior performance of the proposed method.
Researcher Affiliation Academia Jianwei Li1*, Zitong Yu2*, Jingang Shi1 1School of Software Engineering,Xi an Jiaotong University 2Great Bay University
Pseudocode No The paper describes the model architecture and its components (PFE, TFA) with figures and mathematical formulations, but does not include structured pseudocode or algorithm blocks.
Open Source Code Yes The codes are available at https://github.com/LJWGIT/Arbitrary Resolution r PPG.
Open Datasets Yes Extensive experiments on three benchmark datasets (UBFC-r PPG, COHFACE and PURE) demonstrate the superior performance of the proposed method.
Dataset Splits Yes We first conduct experiments of r PPG-based HR measurement on three benchmark datasets with their original protocols and normal setting.
Hardware Specification Yes The proposed method is trained with batchsize 2 on RTX3090 GPU with Py Torch. ... Considering the deployment on edge devices, we provide a detailed analysis of complexity as well as the inference times on different devices (i.e., Nvidia RTX 3090 and Jetson AGX Orin) in Table 4.
Software Dependencies No The paper mentions 'Py Torch' but does not specify its version number or any other software dependencies with version numbers.
Experiment Setup Yes The proposed method is trained with batchsize 2 on RTX3090 GPU with Py Torch. The Adam optimizer is used and the learning rate is set as 1e-4. The weight decay is 5e-5.