动态调整手势检测模型，手机端低功耗高实时
Scale-Gest: Scalable Model-Space Synthesis and Runtime Selection for On-Device Gesture Detection
- 构建可扩展的微型YOLO模型族，支持运行时灵活切换
- 实测每帧能耗降为1.6毫焦（原6.9毫焦），延迟仅6毫秒
- 适配不同电量与性能需求，适合车载等移动场景
在严苛的实时性、功耗和内存约束下实现移动端基于机器学习的手势检测极具挑战，尤其面对电池电量不一的移动设备。现有边缘AI部署多依赖单一固定检测器，限制了优化空间。本文提出Scale-Gest，一种运行时自适应手势检测框架，将检测器空间扩展为密集的tiny-YOLO架构族。通过分析不同模型分辨率-步幅工作点，引入多个设备校准的ACE（准确率-复杂度-能耗）配置。轻量级运行时控制器在用户定义及电池约束下选择合适的ACE模式，同时采用运动感知的手势追踪感兴趣区域（ROI）门，降低输入复杂度。为评估系统在真实驾驶场景下的表现，我们引入一个带时间标注的驾驶员模拟手势数据集DSG-18。相比单检测器方法，Scale-Gest在保持事件级F1值的同时显著降低能耗与延迟：在电池供电笔记本上运行手势流时，每帧能耗降低4倍（从6.9 mJ降至1.6 mJ），事件级F1维持在0.8–0.9，平均延迟低至6毫秒。
原文摘要 · Abstract (English)
Realizing on-device ML-based gesture detection under tight real-time performance, energy and memory constraints is challenging, especially when considering mobile devices with varying battery-power levels. Existing EdgeAI deployments typically rely on a single fixed detector, limiting optimization opportunities. We present Scale-Gest, a novel run-time adaptive gesture detection framework that expands the detector space into a dense family of tiny-YOLO architectures. We introduce multiple novel device-calibrated ACE (Accuracy-Complexity-Energy) profiles by analyzing different model-resolution-stride operating points. A lightweight run-time controller selects an appropriate ACE mode under user-defined and battery constraints, while a motion-aware hand-gesture-tracking ROI gate crops the input for reduced complexity detection. To evaluate performance of our system in real-world car driving scenarios, we introduce a temporally-annotated Driver Simulated Gesture (DSG-18) dataset. Scale-Gest maintains event-level F1 while significantly reducing energy and latency compared to single-detector approaches. On a battery-powered laptop running gesture streams, our ACE controller reduces per-frame energy by 4x (from 6.9 mJ to 1.6 mJ) while maintaining high gesture-detection performance (event-level F1 = 0.8-0.9) and low mean latency (6 ms).
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。