基于坐标注意力脉冲神经网络的注视估计方法OA北大核心CSTPCD
Gaze Estimation Method Based on Coordinate Attention and Spiking Neural Network
针对传统相机在拍摄人眼运动时易产生动态模糊、时间分辨率低等问题,采用事件相机近眼拍摄构建Spiking-Eye数据集,并提出一种坐标注意力的脉冲神经网络模型(CA-SpikingRepVGG).模型读取编码后的事件数据,经过带坐标注意力的主干网络进行特征提取,最后馈入检测头进行检测.实验结果显示:CA-SpikingRepVGG的平均检测精确率RP 达到了70.8%,与SpikingVGG-16 比较,该模型的RP 提高了 15.9%,召回率Rr…查看全部>>
The problems of dynamic blur and low temporal resolution in capturing eye movements with traditional cameras are addressed by employing an event camera for close-range capture and constructing a spiking-eye dataset.A spiking neural network model with a coordinate attention referred to as CA-SpikingRepVGG.The model reads encoded event data and performs feature extraction using the attention-based backbone network,followed by detection using the de…查看全部>>
王红霞;赵志国
沈阳理工大学,辽宁 沈阳 110158沈阳理工大学,辽宁 沈阳 110158
机器视觉目标检测脉冲神经网络注视估计坐标注意力召回率事件相机
machine visionobject detectionspiking neural networkgaze estimationcoordinate attentionrecallevent camera
《计量学报》 2024 (7)
982-988,7
辽宁省自然科学基金(2022-MS-276)
评论