In general, facial image-based remote photoplethysmography (rPPG) methods use color-based and patch-based region-of-interest (ROI) selection methods to estimate the blood volume pulse (BVP) and beats per minute (BPM). Anatomically, the thickness of the skin is not uniform in all areas of the face, so the same diffuse reflection information cannot be obtained in each area. In recent years, various studies have presented experimental results for their ROIs but did not provide a valid rationale for the proposed regions. In this paper, to see the effect of skin thickness on the accuracy of the rPPG algorithm, we conducted an experiment on 39 anatomically divided facial regions. Experiments were performed with seven algorithms (CHROM, GREEN, ICA, PBV, POS, SSR, and LGI) using the UBFC-rPPG and LGI-PPGI datasets considering 29 selected regions and two adjusted regions out of 39 anatomically classified regions. We proposed a BVP similarity evaluation metric to find a region with high accuracy. We conducted additional experiments on the TOP-5 regions and BOT-5 regions and presented the validity of the proposed ROIs. The TOP-5 regions showed relatively high accuracy compared to the previous algorithm’s ROI, suggesting that the anatomical characteristics of the ROI should be considered when developing a facial image-based rPPG algorithm.

This study presents a novel hybrid deep learning model for arrhythmia classification from electrocardiogram signals, utilizing the stockwell transform for feature extraction. As ECG signals are time-series data, they are transformed into the frequency domain to extract relevant features. Subsequently, a CNN is employed to capture local patterns, while a transformer architecture learns long-term dependencies. Unlike traditional CNN-based models that require R-peak detection, the proposed model operates without it and demonstrates superior accuracy and efficiency. The findings contribute to enhancing the accuracy of ECG-based arrhythmia diagnosis and are applicable to real-time monitoring systems. Specifically, the model achieves an accuracy of 97.8% on the Icentia11k dataset using four arrhythmia classes and 99.58% on the MIT-BIH dataset using five arrhythmia classes.

In this paper, we propose a fast Motion Estimation (ME) algorithm with Adaptive Search Range (ASR) decision to further accelerate the Graphics Processing Units (GPU)-based ME for High Efficiency Video Coding (HEVC). The proposed approach adaptively decides search ranges on the Central Processing Unit (CPU) and transfers them to the GPU. Then, the GPU performs ME process in parallel. The proposed approach solves the dependency problem in the Motion Vector Predictor (MVP) derivation stage by using only temporal Motion Vectors (MVs). The proposed algorithm yields the total encoding time reduction of 40.3% with negligible Rate Distortion (RD) loss of 1.2%. In terms of ME, the GPU-based ME with ASR decision provides the time reduction of 54.7% and 1446.8× speed-up on average compared to the GPU-based ME without ASR decision and the full-search ME in the reference model, respectively.

In this paper, we present a novel scheduling solution for a class of System-on-Chip (SoC) systems where heterogeneous chip resources (DSP, FPGA, GPU, etc.) must be efficiently scheduled for continuously arriving hierarchical jobs with their tasks represented by a directed acyclic graph. Traditionally, heuristic algorithms have been widely used for many resource scheduling domains, and Heterogeneous Earliest Finish Time (HEFT) has been a dominating state-of-the-art technique across a broad range of heterogeneous resource scheduling domains over many years. Despite their long-standing popularity, HEFT-like algorithms are known to be vulnerable to a small amount of noise added to the environment. Our Deep Reinforcement Learning (DRL)-based SoC Scheduler (DeepSoCS), capable of learning the “best” task ordering under dynamic environment changes, overcomes the brittleness of rule-based schedulers such as HEFT with significantly higher performance across different types of jobs. We describe a DeepSoCS design process using a real-time heterogeneous SoC scheduling emulator, discuss major challenges, and present two novel neural network design features that lead to outperforming HEFT: (i) hierarchical job- and task-graph embedding; and (ii) efficient use of real-time task information in the state space. Furthermore, we introduce effective techniques to address two fundamental challenges present in our environment: delayed consequences and joint actions. Through an extensive simulation study, we show that our DeepSoCS exhibits the significantly higher performance of job execution time than that of HEFT with a higher level of robustness under realistic noise conditions. We conclude with a discussion of the potential improvements for our DeepSoCS neural scheduler.

Deep learning technology has developed constantly and is applied in many fields. In order to correctly apply deep learning techniques, sufficient learning must be preceded. Various conditions are necessary for sufficient learning. One of the most important conditions is training data. Collecting sufficient training data is fundamental, because if the training data are insufficient, deep learning will not be done properly. Many types of training data are collected, but not all of them. So, we may have to collect them directly. Collecting takes a lot of time and hard work. To reduce this effort, the data augmentation method is used to increase the training data. Data augmentation has some common methods, but often requires different methods for specific data. For example, in order to recognize sign language, video data processed with openpose are used. In this paper, we propose a new data augmentation method for sign language data used for learning translation, and we expect to improve the learning performance, according to the proposed method.

The photoplethysmography (PPG) signal contains various information that is related to CVD (cardiovascular disease). The remote PPG (rPPG) is a method that can measure a PPG signal using a face image taken with a camera, without a PPG device. Deep learning-based rPPG methods can be classified into three main categories. First, there is a 3D CNN approach that uses a facial image video as input, which focuses on the spatio-temporal changes in the facial video. The second approach is a method that uses a spatio-temporal map (STMap), and the video image is pre-processed using the point where it is easier to analyze changes in blood flow in time order. The last approach uses a preprocessing model with a dichromatic reflection model. This study proposed the concept of an axis projection network (APNET) that complements the drawbacks, in which the 3D CNN method requires significant memory; the STMap method requires a preprocessing method; and the dyschromatic reflection model (DRM) method does not learn long-term temporal characteristics. We also showed that the proposed APNET effectively reduced the network memory size, and that the low-frequency signal was observed in the inferred PPG signal, suggesting that it can provide meaningful results to the study when developing the rPPG algorithm.