研究人员开发了PI-TTA，一个用于移动设备上鲁棒的人类活动识别的新框架。该方法解决了在适应真实世界传感器数据时遇到的挑战，例如旋转和采样率漂移，这些问题可能导致标准的自适应方法不稳定。PI-TTA使用物理一致性约束，如重力一致性和时间连续性，来稳定在线更新，使其适用于具有最小开销的设备部署。实验表明，在多个数据集上，准确性显著提高，物理违规行为减少。 AI
影响 增强了移动传感应用中设备上AI的可靠性和准确性。
排序理由 介绍特定AI任务新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
Source-free test-time adaptation (TTA) is appealing for mobile and wearable sensing because it enables on-device personalization from unlabeled test streams without centralizing private data. However, sensor-based human activity recognition (HAR) poses challenges that are less pr…
Source-free test-time adaptation (TTA) is appealing for mobile and wearable sensing because it enables on-device personalization from unlabeled test streams without centralizing private data. However, sensor-based human activity recognition (HAR) poses challenges that are less pr…