This paper introduces UrbAM-ReID, a new long-term geo-positioned urban ReID dataset. It is composed by four sub-datasets recording the same trajectory at the UAM Campus, each one recorded in different seasons and incl...
详细信息
ISBN:
(纸本)9798350349405;9798350349399
This paper introduces UrbAM-ReID, a new long-term geo-positioned urban ReID dataset. It is composed by four sub-datasets recording the same trajectory at the UAM Campus, each one recorded in different seasons and including an inverse direction recording. While most of the current datasets in the state-of-the-art focus on person re-identification, with vehicles as the second most explored object, our work specifically addresses urban objects re-identification, currently, waste containers, rubbish bins, and crosswalks. The dataset provides different attributes of the annotated objects, like their classes, their foreground or background status and the geo-position. Several evaluation configurations can be defined to simulate realistic scenarios that may arise in actual situations within the management of urban elements, considering the utilization of just visual data, or incorporating additional attributes, providing different complexity levels. Finally, the dataset is used for defining a benchmark where two state-of-the-art systems are evaluated. The dataset and supplementary material is available in https://***/vpulab/UrbAM-ReID
Digital signalprocessing (DSP) technology is a technology that completes data statistics and processing through digital computing, and is widely used in communication, measurement, statistics and other technical fiel...
详细信息
Drive testing plays a crucial role in assessing and optimizing the performance of mobile networks for operators. With the transition from fourth-generation (4G) to fifth-generation (5G) networks, it becomes essential ...
详细信息
ISBN:
(纸本)9798350384826;9798350384819
Drive testing plays a crucial role in assessing and optimizing the performance of mobile networks for operators. With the transition from fourth-generation (4G) to fifth-generation (5G) networks, it becomes essential to evaluate and compare the performance of these technologies in real-world scenarios. Drive test real measurement data assumes a central role in optimizing mobile networks. Operators gain invaluable insights into critical factors such as signal quality, data throughput, and handover events by conducting physical drive tests. This empirical data empowers them to identify areas with inadequate coverage, optimize antenna settings, refine handover parameters, and strategically deploy additional base stations. This paper provides symmetrical performance analyses based on an experimental study of operating 5G cellular networks in Muscat, Oman. The data measurement campaign was collected using mobile devices from three mobile network providers (Omantel, Ooredoo, and Vodafone). The network's performance was evaluated using various metrics, including channel quality indicators, data rates and handover events. Assessment results show that the leading networks in the measured area are 5G networks for all mobile network providers. While 5G has showcased incremental improvements over 4G in some instances, it's still in its nascent phase. Recommendations include continuous network monitoring by operators and augmented 5G base station deployment to heighten the quality of service by ensuring higher data rates and minimal latency.
The combination of electromyography (EMG) and electroencephalography (EEG) sensors with Internet of Things (IoT) enabled robots has marked a turning point in human-robot interaction. This chapter examines the groundbr...
详细信息
Recent advancements in deep learning have signif-icantly enhanced the rapid and precise classification of medical images. vision transformers, an advanced model, have started replacing CNN s in several medical image t...
详细信息
Accurate gait phase recognition is crucial for real-time analysis and intervention in rehabilitation, biomechanics, and prosthetics. However, achieving this is challenging due to the diverse machine learning (ML) trai...
详细信息
ISBN:
(纸本)9798350351491;9798350351484
Accurate gait phase recognition is crucial for real-time analysis and intervention in rehabilitation, biomechanics, and prosthetics. However, achieving this is challenging due to the diverse machine learning (ML) training methods. This study employs ML algorithms to classify gait phases, focusing on stance and swing phases, utilizing open-source data from 100 participants (41.91 +/- 5.3 yrs). The classification algorithms considered are k-Nearest Neighbor (k-NN), Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and Naive Bayesian (NB) algorithms. The study evaluates these algorithm performances using two training methods with five and ten lower body movements: one randomly selects 80% of stance and swing phase data for training, while the other divides data by participants, allocating 80% for training and 20% for testing. When assessing accuracy with five movements, RF achieved 99.8% for both training methods. With ten movements, RF achieved a high accuracy of 99.9% using the second method. Notably, the performance of all ML algorithms exhibited improvement when considering data from ten movements versus five. Additionally, it was observed that the second training method proved more effective with five-movement data compared to data involving ten movements. This comprehensive evaluation highlights the potential of machine learning algorithms for accurate gait phase recognition in diverse applications with varied training methods.
The proceedings contain 145 papers. The topics discussed include: design and implementation of mimic defense gateway;attenuation correction of CT-guided PET images based on deep convolutional neural networks;low-bit h...
ISBN:
(纸本)9798350350920
The proceedings contain 145 papers. The topics discussed include: design and implementation of mimic defense gateway;attenuation correction of CT-guided PET images based on deep convolutional neural networks;low-bit hybrid RIS enhanced beamforming design with imperfect CSI;channel estimation algorithm based on multi-modal neural network;an active visual perception methodology for the safety of human-robot collaborations;radar forward-looking super-resolution imaging based on joint projection and non-convex two-step regularization;invisible and multi-triggers backdoor attack approach on deep neural networks through frequency domain;establishment and research of liver medical image online database based on deep learning;and aerial target classification using micro-doppler spectrum based on PCT and ResNet34.
RF signals can be leveraged for many sensing and monitoring tasks in industrial, home, or robot applications. Despite the advantages of leveraging WiFi sensing modality, no versatile WiFi sensors are available. We dev...
详细信息
The development of artificial intelligence (AI) assistants has revolutionized human-machine interaction, but traditional systems often rely on singular input modalities such as voice commands. This project, AURA: AI R...
详细信息
In this paper, the automatic assembly of shaft-hole with robot is researched, and the accuracy range of robot shaft hole assembly based on image recognition algorithm is explored. According to the accuracy experiment ...
详细信息
ISBN:
(纸本)9798350389814;9798350389807
In this paper, the automatic assembly of shaft-hole with robot is researched, and the accuracy range of robot shaft hole assembly based on image recognition algorithm is explored. According to the accuracy experiment requirements, the robot system and shaft-hole assembly platform are built, the eye-in-hand system calibration scheme is designed. A recognition algorithm combining image preprocessing technology, edge detection and Hough transform circle detection is proposed to locate the mounting hole and realize the shaft-hole assembly with the robot. In order to improve the image recognition process, image preprocessing technologies such as gray level processing, Gaussian filter are used to reduce the influence of the surrounding environment and noise points. Edge detection with Sobel operator is adopted to enhance the target features of the image and improve the accuracy of circle center recognition. Hole center coordinates is obtained by Hough transform circle detection. Finally, the hand-eye relationship obtained by calibration is used to drive the robot to realize the shaft-hole assembly. In order to obtain the accuracy range of shaft-hole assembly, several experiments were carried out, and the experimental results showed that the error of shaft-hole assembly is limited to +/- 1 mm.
暂无评论