Abstract: With the development of sensing technologies and machine learning, electroencephalography (EEG) techniques that can identify human emotions and inner states through physiological signals have been actively developed and applied to automobiles, robotics, healthcare, and…
Abstract: Traffic state estimation (TSE) is complicated by sensor noise and data sparsity. This paper introduces physics-informed deep learning (PIDL) to solve this problem. PIDL strengthens a deep learning neural network to better estimate traffic…
Abstract: We test a Deep Learning model for 3D medical image segmentation. Our model addresses manual segmentation’s high inter-rater contouring variability and contouring time. The careful analysis can be applied to other medical image segmentation…
Abstract: Deep learning-based automatic recognition of the scanning or exposing region in medical imaging automation is a promising new technique that can reduce radiographer workload, optimize imaging workflow, and improve image quality. X-ray imaging research…
Abstract: Federated learning allows deep learning model training without sharing patient data. Federated learning models may perform poorly due to heterogeneous data across institutions. We propose SplitAVG, a heterogeneity-aware federated learning method, to overcome performance…
Abstract: Due to deployment, coverage, and capacity issues, ground wireless communication networks cannot support AI applications like intelligent transportation systems (ITS). The industry is researching the space-air-ground integrated network (SAGIN). SAGIN is flexible, reliable, and…
Abstract: Transportation services are increasingly important in people’s daily lives and benefit individuals and economic development. Traffic flows are unpredictable and volatile, limiting transportation services. To stabilize intelligent transport systems and optimize traffic scheduling, precise…
Abstract: Autonomous driving and ADAS tasks are growing. However, porting multiple functions into a power-constrained computing device is difficult. This work aims to simplify the pixel-wise driving scene understanding learning process. This paper uses semantic…
Abstract: A real-time tracking algorithm using improved convolutional neural network (CNN) and transfer learning was proposed to improve image feature extraction, tracking time, and accuracy of traditional real-time algorithms for aerial vehicle traffic tracking. For…
Abstract: Facial phenotyping for medical prediagnosis has been successfully used to preclinically assess a variety of rare genetic diseases. Facial biometrics has rich links to underlying genetic or medical causes. This paper extends facial prediagnosis…
Abstract: To optimize on-street parking resources and create effective parking policies, parking occupancy data is essential. Advanced and expensive occupancy monitoring technologies often provide such data. Integrating bay-level occupancies and parking data from other systems…
Abstract: Noise and missing data corrupt medical data. In medical settings, missing patterns occur in bursts due to sensors being off or data being collected unevenly. Sequential variational autoencoders (VAEs) are proposed to model heterogeneous…
WhatsApp us