用头戴式传感器高效识别高低层级动作，模型小巧可直接部署。
EgoCHARM: Resource-Efficient Hierarchical Activity Recognition using an Egocentric IMU Sensor
- 分层半监督学习，仅需高层动作标签训练
- 高层与低层动作识别F1分别达0.826和0.855
- 模型参数仅63k/22k，适合嵌入式实时运行
智能眼镜上的姿态识别（HAR）在健康追踪和上下文感知助手中有广泛应用。然而，现有头戴式动作识别方法普遍存在性能差或资源消耗高的问题。本文提出一种资源高效（内存、计算、功耗、样本）的机器学习算法EgoCHARM，仅使用单个头戴式惯性测量单元（IMU）即可识别高阶与低阶动作。该分层算法采用半监督学习策略，主要依赖高阶动作标签进行训练，从而学习可泛化的低阶运动表征，有效用于低阶动作识别。我们在9个高阶和3个低阶动作上进行评估，高阶与低阶动作识别的F1分数分别为0.826和0.855，模型参数仅需63k（高阶）和22k（低阶），使低阶编码器可直接部署于当前IMU芯片的计算单元。最后，我们通过敏感性分析提供了结果与洞察，揭示了头戴式IMU进行动作识别的机遇与局限。
原文摘要 · Abstract (English)
Human activity recognition (HAR) on smartglasses has various use cases, including health/fitness tracking and input for context-aware AI assistants. However, current approaches for egocentric activity recognition suffer from low performance or are resource-intensive. In this work, we introduce a resource (memory, compute, power, sample) efficient machine learning algorithm, EgoCHARM, for recognizing both high level and low level activities using a single egocentric (head-mounted) Inertial Measurement Unit (IMU). Our hierarchical algorithm employs a semi-supervised learning strategy, requiring primarily high level activity labels for training, to learn generalizable low level motion embeddings that can be effectively utilized for low level activity recognition. We evaluate our method on 9 high level and 3 low level activities achieving 0.826 and 0.855 F1 scores on high level and low level activity recognition respectively, with just 63k high level and 22k low level model parameters, allowing the low level encoder to be deployed directly on current IMU chips with compute. Lastly, we present results and insights from a sensitivity analysis and highlight the opportunities and limitations of activity recognition using egocentric IMUs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。