[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fcwrkUydRYH9vgIhCoyICAsQ4-zezlwcMHCd1YbYIjhg":3},{"answer":4,"createTime":5,"id":6,"options":7,"origin":12,"question":19,"related":20,"source":30,"type":31},[],"2025-12-09 14:36:49",257147087,[8,9,10,11],"1470","1960","980","1078",{"count":13,"courseId":14,"courseImg":15,"courseName":16,"workId":17,"workName":18},21,"ab3315e1f5c5386b6905fd3d47ad00c6","https:\u002F\u002Ftihai-oss-cloud.itihey.com\u002Fimg\u002Fcff6dd68d0e399d249aaba4ab132f3e2.png","神经网络与深度学习","work_48103202","第10次课作业-part1(客观题):卷积神经网络(四)——一阶段目标检测","现使用YOLO v1网络进行目标检测,待检测的物体种类为20种,输入图像被划分成7*7个格子,每个格子生成2个候选框,则YOLO v1网络最终的全连接层输出维度为()",[21,32,41,50,53,62,71,80,89,98],{"answer":22,"createTime":5,"id":23,"options":24,"question":29,"source":30,"type":31},[],257147082,[25,26,27,28],"R-CNN","YOLOV3","SPP-net","Faster-rcnn","下列目标检测网络中,哪个是一阶段的网络( )","v1",0,{"answer":33,"createTime":5,"id":34,"options":35,"question":40,"source":30,"type":31},[],257147084,[36,37,38,39],"多尺度特征融合","锚点框(Anchor Boxes)","更深的网络结构","批量归一化(Batch Normalization)","YOLOv3相比YOLOv2引入了哪项关键改进",{"answer":42,"createTime":5,"id":43,"options":44,"question":49,"source":30,"type":31},[],257147085,[45,46,47,48],"只在一个尺度上进行目标检测","只使用单个GPU进行训练和测试","只需要一次前向传播就可以完成目标检测任务","只使用单个卷积核进行检测","SSD算法中的"Single Shot"表示什么",{"answer":51,"createTime":5,"id":6,"options":52,"question":19,"source":30,"type":31},[],[8,9,10,11],{"answer":54,"createTime":5,"id":55,"options":56,"question":61,"source":30,"type":31},[],257147088,[57,58,59,60],"使用了更深的网络结构Darknet-53","引入了anchor机制来预测边界框","采用了FPN(特征金字塔网络)来增强多尺度目标检测","使用了ROI Pooling层进行区域分类","以下哪个不是YOLOv3算法的特点",{"answer":63,"createTime":5,"id":64,"options":65,"question":70,"source":30,"type":31},[],257147091,[66,67,68,69],"只计算分类损失","只计算定位损失","分类损失和定位损失的加权和","分类损失、定位损失和置信度损失的加权和","5. YOLOv1算法中,如何计算损失函数",{"answer":72,"createTime":5,"id":73,"options":74,"question":79,"source":30,"type":31},[],257147094,[75,76,77,78],"删除非极大值抑制步骤(NMS)","适当增加网格的大小S","适当增加每个网格的边界框的数量","样本增强","对于YOLO v1算法,下面哪种做法可能不能增加其检测精度()",{"answer":81,"createTime":5,"id":82,"options":83,"question":88,"source":30,"type":31},[],257147096,[84,85,86,87],"通过在不同层级上提取特征并生成预测","通过特征金字塔网络","通过不同分辨率的输入图像","通过不同大小的卷积核","SSD算法中的多尺度检测是如何实现的",{"answer":90,"createTime":5,"id":91,"options":92,"question":97,"source":30,"type":31},[],257147097,[93,94,95,96],"高精度","高速度","适用于小目标检测","适用于多尺度目标","YOLO(You Only Look Once)目标检测算法的主要优点是什么",{"answer":99,"createTime":5,"id":100,"options":101,"question":106,"source":30,"type":31},[],257147099,[102,103,104,105],"YOLOv2","YOLOv1","YOLOv3","YOLOv4","YOLO系列算法中的哪个版本引入了多尺度特征图来提高对不同大小目标的检测性能"]