number9473 / nn-algorithm

algorithm for neural network
258 stars 59 forks source link

SSD: Single Shot MultiBox Detector #143

Open joyhuang9473 opened 7 years ago

joyhuang9473 commented 7 years ago

SSD: Single Shot MultiBox Detector

joyhuang9473 commented 7 years ago

How to train your own Object Detector with TensorFlow’s Object Detector API

https://medium.com/towards-data-science/how-to-train-your-own-object-detector-with-tensorflows-object-detector-api-bec72ecfe1d9

joyhuang9473 commented 7 years ago

Tensorflow Object Detection API

https://github.com/tensorflow/models/tree/master/object_detection

joyhuang9473 commented 7 years ago

SSD论文阅读(Wei Liu——【ECCV2016】SSD Single Shot MultiBox Detector)

https://www.bbsmax.com/A/gGdX0Vepd4/

joyhuang9473 commented 7 years ago

SSD:Single Shot MultiBox Detector代码解读

http://blog.csdn.net/u011956147/article/details/73028773

joyhuang9473 commented 7 years ago

Train SSD on custom dataset

https://github.com/weiliu89/caffe/wiki/Train-SSD-on-custom-dataset

joyhuang9473 commented 7 years ago

用SSD的物体检测网络,它的label是怎么写入lmdb的

http://caffecn.cn/?/question/831

joyhuang9473 commented 7 years ago

SSD: Signle Shot Detector 用于自然场景文字检测

joyhuang9473 commented 7 years ago

Ubuntu上用caffe的SSD方法训练umdfaces数据集

http://blog.csdn.net/u013738531/article/details/61934587

joyhuang9473 commented 7 years ago

shufflenet-based SSD caffemodel:

@ananddb90 @farmingyard

joyhuang9473 commented 6 years ago

follow the ncnn's issue "Support which detection network model,Faster-RCNN or SSD or YOLO?" https://github.com/Tencent/ncnn/issues/152#issuecomment-334775869

joyhuang9473 commented 6 years ago

mobilenet-based SSD caffemodel:

https://github.com/chuanqi305/MobileNet-SSD

73

dlyldxwl commented 6 years ago

Thanks you for your job ! I'm interested in shufflenet-ssd,but the link of shufflenet-ssd caffemodel is not effective now. Can you share with me the caffemodel and train or deploy prototxt? My email is 844518133@qq.com. Thanks again !

408550969 commented 6 years ago

楼主你好,请问 shufflenet-based SSD 在VOC上的精度能达到多少?