📄 中文摘要
日常活动的识别是有效环境辅助生活（AAL）系统的重要组成部分，尤其是在室内环境中监测老年人的健康和支持其独立性。然而，开发稳健的活动识别系统面临诸多挑战，包括类内变异、类间相似性、环境变化、摄像机视角以及场景复杂性。该研究提出了一种多模态方法，旨在识别老年人在AAL环境中的日常生活活动。所提出的系统将通过3D卷积神经网络（CNN）处理的视觉信息与通过图卷积网络分析的3D人体姿态数据相结合，利用上下文信息增强活动识别的准确性。
通过多模态深度学习识别日常活动：一种面向环境辅助生活的视频、姿态和物体感知方法
📄 English Summary
Recognition of Daily Activities through Multi-Modal Deep Learning: A Video, Pose, and Object-Aware Approach for Ambient Assisted Living
Recognition of daily activities is a crucial component of effective Ambient Assisted Living (AAL) systems, particularly for monitoring the well-being and supporting the independence of older adults in indoor environments. Developing robust activity recognition systems encounters significant challenges such as intra-class variability, inter-class similarity, environmental variability, camera perspectives, and scene complexity. This study proposes a multi-modal approach for recognizing activities of daily living tailored for older adults within AAL settings. The proposed system integrates visual information processed by a 3D Convolutional Neural Network (CNN) with 3D human pose data analyzed by a Graph Convolutional Network, leveraging contextual information to enhance the accuracy of activity recognition.