Scholarlink

<p>Falls pose a significant threat to human safety, making rapid and accurate detection and response essential. Time Exploration Network (TExNet), an attention-enhanced network tailored for identifying falls accurately and rapidly, is proposed in this paper. Unlike existing studies that rely on simulated environments, TExNet addresses the gap between simulated and real scenarios by integrating multi-branch timing and classification characteristics. It features a two-branch adaptive fusion framework, leveraging Convoluational Neural Network (CNN) and Transformer architectures to capture both local and global dependencies effectively. Additionally, dual-branch adaptive fusion framework incorporates dilated convolution and time series positional decomposition to enhance temporal correlation understanding. To handle data distribution variations, it employs an Invariant Risk Minimization (IRM) inspired loss function, penalizing misclassification of positive examples. This reduces model reliance on specific environments and improves action understanding. Moreover, a data self-conditioning module enhances data diversity and tackles imbalance issues. The model is deployed using a fine-tuning strategy based on a pre-trained framework combined with few-shot learning for downstream tasks. Experimental results show that the fine-tuned model achieves a recall of 92.16%, indicating its strong ability to rapidly adapt to new data distributions. Extensive experiments also validate the superiority of TExNet compared with existing approaches.</p> Graphical abstract <p>Given the significant importance of fall detection in medical health monitoring, this paper proposes a fall detection method based on Transformer and dilated convolutional neural network. By studying the movement data of the wrist, the essential features of fall detection can be learned. Use a large-scale publicly available dataset for pre-training, and fine-tune on a small dataset collected by self-developed hardware to weaken the dependence of fall recognition on the environment. The simulation recall rate reached 92.16%.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

TExNet: long short-term fast fall detection based on attention enhancement and self-adaption

  • 期刊论文

  • 发表日期：

    2026年6月22日

查看全文
  • Yixin Ding,
  • Baoxuan Fang,
  • Ruihan Gao,
  • Xunhe Yin

摘要

Falls pose a significant threat to human safety, making rapid and accurate detection and response essential. Time Exploration Network (TExNet), an attention-enhanced network tailored for identifying falls accurately and rapidly, is proposed in this paper. Unlike existing studies that rely on simulated environments, TExNet addresses the gap between simulated and real scenarios by integrating multi-branch timing and classification characteristics. It features a two-branch adaptive fusion framework, leveraging Convoluational Neural Network (CNN) and Transformer architectures to capture both local and global dependencies effectively. Additionally, dual-branch adaptive fusion framework incorporates dilated convolution and time series positional decomposition to enhance temporal correlation understanding. To handle data distribution variations, it employs an Invariant Risk Minimization (IRM) inspired loss function, penalizing misclassification of positive examples. This reduces model reliance on specific environments and improves action understanding. Moreover, a data self-conditioning module enhances data diversity and tackles imbalance issues. The model is deployed using a fine-tuning strategy based on a pre-trained framework combined with few-shot learning for downstream tasks. Experimental results show that the fine-tuned model achieves a recall of 92.16%, indicating its strong ability to rapidly adapt to new data distributions. Extensive experiments also validate the superiority of TExNet compared with existing approaches.

Graphical abstract

Given the significant importance of fall detection in medical health monitoring, this paper proposes a fall detection method based on Transformer and dilated convolutional neural network. By studying the movement data of the wrist, the essential features of fall detection can be learned. Use a large-scale publicly available dataset for pre-training, and fine-tune on a small dataset collected by self-developed hardware to weaken the dependence of fall recognition on the environment. The simulation recall rate reached 92.16%.