低功耗高精度的手势识别算法实时手势识别准确率测试数据

实时手势识别准确率主要面向高能效人机交互芯片技术中低功耗高性能手势识别芯片的算法研究，基于复旦大学设计、流片后的芯片测试产生，主要记录了在不同手势移动速度、不同手势距离摄像头距离下产生的识别效果观测值，数据量。实时手势识别准确率对应于手势识别芯片的检测实时性和准确率，是手势识别芯片最重要的指标之一，体现手势识别芯片实用性价值的直观数据。

This dataset targets the algorithm research of low-power, high-performance gesture recognition chips in the field of energy-efficient human-computer interaction chip technology, with real-time gesture recognition accuracy as the core concern. The data is generated from post-fabrication chip tests designed by Fudan University. It mainly records the observed recognition performance values under different gesture moving speeds and different distances between the gesture and the camera, along with the corresponding data volume. Real-time gesture recognition accuracy corresponds to the detection timeliness and accuracy of gesture recognition chips, and is one of the most important metrics for such chips, serving as intuitive data reflecting the practical value of gesture recognition chips.

搜集汇总

背景与挑战

背景概述

该数据集是用于低功耗高精度手势识别算法研究的测试数据，基于复旦大学设计并流片后的芯片测试产生，记录了在不同手势移动速度和距离下的实时识别准确率观测值，以评估芯片的实用性和性能。数据量较小（472.96KB），属于集成电路技术领域，关联国家重点研发计划项目'高能效人机交互芯片技术'。