Chen Ma

Results 48 comments of Chen Ma

@nils489 @shijx12 I download and compile your caffe-fast-rcnn , it seems to compile success. But when I run demo.py, I found that your src/caffe/proto/caffe.proto **did not even have roi_pooling_param?** Did...

@Xiaocong6 正在开发,预计12月多开发完毕。其实chainer和PyTorch很相似的。

@Supltz @Xiaocong6 我最近有时间开始开发PyTorch版本了 ,敬请期待

@Gepeng18 最主要是`FaceMaskCropper.get_cropface_and_box`, 其实在config.py里面定义了精细的区域划分规则(利用了哪几个landmark point)以及AU group的规则。

@Gepeng18 我做好后会通知你的,其实现在有个半成品,但是因为在忙另一个会议,你要的话半成品的pytorch代码可以发你。

It looks like that the python's project PATH does not configured correctly. For example , you may modify https://github.com/machanic/AU_R-CNN/blob/master/AU_rcnn/train.py#L4 to the AU R-CNN project's absolute path in your machine.

没听懂,但是你如果能把ROI对准五官各个区域应该也是可以的,你可以实验一下,关键是看ROI能否对准脸上的部位。

All the related works report the average F1 score, rather than average. Because the accuracy is not fair in the case that **the data distribution over different AU classes are...