helpResearch questionHAR模型鲁棒性、泛化与数据增强
基于多任务预训练和对比学习机制，Transformer模型在活动识别和情感检测等应用中能否超越传统ResNet架构？UbiComp '23Transformer-based LLMs for Sensor Data
传感、识别与安全 / HAR模型鲁棒性、泛化与数据增强
基于多任务预训练和对比学习机制，Transformer模型在活动识别和情感检测等应用中能否超越传统ResNet架构？
Similar questions
helpResearch questionHAR模型鲁棒性、泛化与数据增强
Kolmogorov-Arnold网络（KANs）能否更高效地对IMU传感器数据进行特征提取以支持人类活动识别？UbiComp '24Initial Investigation of Kolmogorov-Arnold Networks (KANs) as Feature Extractors for IMU Based Human Activity Recognition
helpResearch questionHAR模型鲁棒性、泛化与数据增强
单通道、多通道与跨通道特征对KAN架构在人类活动识别中的表现有何影响？UbiComp '24Initial Investigation of Kolmogorov-Arnold Networks (KANs) as Feature Extractors for IMU Based Human Activity Recognition
helpResearch questionHAR模型鲁棒性、泛化与数据增强
在相比CNN的特征提取中，KAN在参数效率和分类准确性上是否具有优势？UbiComp '24Initial Investigation of Kolmogorov-Arnold Networks (KANs) as Feature Extractors for IMU Based Human Activity Recognition
helpResearch questionHAR模型鲁棒性、泛化与数据增强
主动声学技术能否通过智能眼镜精确监测并识别多样的日常活动？UbiComp '24ActSonic: Recognizing Everyday Activities from Inaudible Acoustic Wave Around the Body
helpResearch questionHAR模型鲁棒性、泛化与数据增强
如何利用深度学习和自监督学习，从无标签数据中提取有效特征以提高活动识别性能？UbiComp '24ActSonic: Recognizing Everyday Activities from Inaudible Acoustic Wave Around the Body
helpResearch questionHAR模型鲁棒性、泛化与数据增强
主动声学传感能否在能耗低、保护隐私的情况下，胜任现有摄像头或惯性传感器无法处理的活动识别任务？UbiComp '24ActSonic: Recognizing Everyday Activities from Inaudible Acoustic Wave Around the Body