H.3.2.2. Computer vision
Fatemeh Naghavi; Kourosh Kiani
Abstract
Monitoring the daily activities of elderly individuals plays a crucial role in accident prevention, health assessment, and improving quality of life. In this paper, we propose a lightweight and efficient convolutional neural network architecture for human activity recognition based on skeletal data. ...
Read More
Monitoring the daily activities of elderly individuals plays a crucial role in accident prevention, health assessment, and improving quality of life. In this paper, we propose a lightweight and efficient convolutional neural network architecture for human activity recognition based on skeletal data. Unlike conventional approaches that rely solely on absolute joint coordinates, the proposed method incorporates short- and long-term frame differences as well as spatial variations across joints to construct complementary views, thereby providing a richer spatiotemporal representation. The architecture consists of multiple convolutional blocks with residual connections, followed by global average pooling and a fully connected layer for final classification. Experimental evaluations conducted on two benchmark datasets, NTU RGB+D and ETRI-Activity3D, demonstrate that while the proposed model may achieve slightly lower accuracy compared to some state-of-the-art methods, it offers high inference speed and low computational complexity. These characteristics make it particularly suitable for real-time applications and deployment on resource-constrained devices, especially in elderly home-care environments.
H.5.9. Scene Analysis
Navid Raisi; Mahdi Rezaei; Behrooz Masoumi
Abstract
Human Activity Recognition (HAR) using computer vision is an expanding field with diverse applications, including healthcare, transportation, and human-computer interaction. While classical approaches such as Support Vector Machines (SVM), Histogram of Oriented Gradients (HOG), and ...
Read More
Human Activity Recognition (HAR) using computer vision is an expanding field with diverse applications, including healthcare, transportation, and human-computer interaction. While classical approaches such as Support Vector Machines (SVM), Histogram of Oriented Gradients (HOG), and Hidden Markov Models (HMM) rely on manually extracted features and struggle with complex motion patterns, deep learning-based models (e.g., Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), Transformer-based models) have improved performance but still face challenges in handling occlusions, noisy environments, and computational efficiency. This paper introduces Attention-HAR, a novel deep neural network model designed to enhance HAR performance through three key innovations: Conv3DTranspose for spatial upsampling, ConvLSTM2D for capturing spatiotemporal patterns, and a custom attention mechanism that prioritizes critical frames within sequences. Unlike conventional attention mechanisms, our approach dynamically assigns weights to key frames, reducing the impact of redundant frames and enhancing interpretability and computational efficiency. Experimental results on the UCF-101 dataset demonstrate that Attention-HAR outperforms state-of-the-art models, achieving an accuracy of 97.61%, a precision of 97.95%, a recall of 97.49%, an F1-score of 97.64, and an AUC of 99.9%. With only 1.26 million parameters, the model is computationally efficient and well-suited for deployment on lightweight platforms. These findings suggest that integrating temporal-spatial feature learning with attention mechanisms can significantly improve HAR in dynamic and complex environments.
