图像目标检测(detection)
jingqiulyue
迭代升级中
展开
-
理解Yolo 的网络参数
终端参数推荐:http://blog.csdn.net/dcrmg/article/details/78565440 训练log中各参数的意义 Region Avg IOU:平均的IOU,代表预测的bounding box和ground truth的交集与并集之比,期望该值趋近于1。 Class:是标注物体的概率,期望该值趋近于1. Obj:期望该值趋近于1. No Obj:期望...原创 2018-03-22 15:32:21 · 1757 阅读 · 0 评论 -
YOLO9000:Better, Faster, Stronger
推荐: http://blog.csdn.net/qq_14845119/article/details/53589282 https://zhuanlan.zhihu.com/p/25052190 https://zhuanlan.zhihu.com/p/25167153?utm_source=wechat_session&utm_medium=social Abstract ...原创 2018-03-19 18:58:22 · 304 阅读 · 0 评论 -
重新训练YOLO v2
推荐: http://blog.csdn.net/u014540717/article/details/53424843 http://blog.csdn.net/hysteric314/article/details/54097845 http://blog.csdn.net/eloise_29/article/details/70215338 Training YOLO v2 on P...原创 2018-03-21 16:40:23 · 1450 阅读 · 5 评论 -
R-FCN:Object Detection via Region-based Fully Convolutional Networks
代码: tensorflow版本: https://github.com/xdever/RFCN-tensorflow https://github.com/alfonsolink/tf_rfcn https://github.com/endernewton/tf-faster-rcnn https://github.com/auroua/tf_rfcn Caffe版本: https...原创 2018-05-04 15:42:40 · 144 阅读 · 0 评论 -
R-FCN caffe 代码运行指南(python3 + anaconda 3)
推荐:http://www.cnblogs.com/peiyuYang/p/7784787.html 1.下载py-RFCN git clone https://github.com/Orpine/py-R-FCN.git 2.下载微软版caffecd $RFCN_ROOT git clone https://github.com/Microsoft/caffe.git3.li...原创 2018-05-06 16:08:36 · 1418 阅读 · 0 评论 -
单类目标检测-yolo
推荐: https://blog.csdn.net/zwx1995zwx/article/details/79901013 https://blog.csdn.net/davidlee8086/article/details/79693079原创 2018-05-13 09:20:34 · 7384 阅读 · 1 评论 -
重新训练R-FCN
运行demo_rfcn.py只是跑人家训练好的模型,接下来自己训练。(VOC数据集) 1.准备数据 下载VOC数据集:# Download the training, validation, test data and VOCdevkitwget http://host.robots.ox.ac.uk/pascal/VOC/voc2007/VOCtrainval_06-Nov-2007....原创 2018-05-06 22:08:34 · 591 阅读 · 0 评论 -
史上最详尽的yolo教程
https://github.com/AlexeyAB/darknet#how-to-train-to-detect-your-custom-objects转载 2018-05-13 16:04:30 · 8788 阅读 · 2 评论