cascade-rcnn
cascade-rcnn copied to clipboard
Caffe implementation of multiple popular object detection frameworks
I haven't run caffe with multiple gpus before, so I'm a little confused that why the batch size can set larger than 1 in Cascade R-CNN given that the input...
loss不下降
我最近用cascade-fpn训练自己的一个数据集,训练这个数据集的时候我把学习率已经调到了0.00001,loss一直在0.4左右不能下降,不知道是什么原因呢?cascade是不是有一定的数据集使用范围?有些数据集并不适用呢·?
why does the bbox_std value differ in different stages, What is this value based on???thanks
How much memory of GPU is needed for inference? I test images with GTX 1060 (6G), error: out of memory!
Hello,@ zhaoweicai, I want to ask you, which file is your cascade loss function, I want to see your loss function, thank you very much.
hi, I want to use res101-15s-800-fpn-cascade to test one image. I want to get boxes and labels, is there any c++ interface to finish it?
I0806 23:44:24.048591 20123 solver.cpp:219] Iteration 9900 (2.14913 iter/s, 46.5305s/100 iters), loss = 0.440841 I0806 23:44:24.048627 20123 solver.cpp:238] Train net output #0: bbox_iou = -1 I0806 23:44:24.048635 20123 solver.cpp:238] Train net...
when I train and test the model for voc data,I always get this problem. 
i trained the alex-9s-600-rpn-base without finetune,here is the snapshort result  but the rpn foreground accuracy is always 0, is it alright?