Dantju comments

Results 29 comments of


                                            Dantju

A new model trained on VGGFace2

@Shahnawazgrewal I add center loss to caffe, I met the error "math_functions.cu:155] Check failed: error == cudaSuccess (11 vs. 0) invalid argument" when I train,have you met this problem or...

Test Problem

@zys1994 i also get worse result in lfw，i want to know how many iter have you train on ResNet?

Test Problem

@happynear hi, have u trained the ResNet using VGG2 dataset? i only get 98.5 on LFW,i want to know how many iter have you train on ResNet using VGG2 dataset?

Downloading is too slow.

why IMDb-Face.csv only has 1048576 images？have u downloaded all dataset?

Get stacked when running test code

I met the same problem，it stop at 》classes ：['_backgroud_','person']

can this code produce pose estimation and person detection result at the same time

how to train pose estamation and person detection at the same time,how to config it

inference time

@leoxiaobin @stefanopini @hellojialee have u test image using HigherHRNet，it seems that it‘s demo results is worser than centernet

inference time

@stefanopini yes in paper HIgherHRNet is better than centernet,but i use centernet and higherHRNet to test same image,it seems that the demo result of centernet is better than HigherHRNet，I confused

关于Z值和keypoints map中亮点所在位置的两个问题

@murdockhou 请问一下有没有考虑过将SPM和CenterNet融合呢，在CenterNet code的基础上添加SPM呢

关于Z值和keypoints map中亮点所在位置的两个问题

@murdockhou 另外我看code是用AI challenger数据集来训练，目前下载不到这个数据集，能否分享这个数据集呢，或者能分享一下在COCO数据集上训练的code吗