unsky

时间过去太久了，有些细节可能有些问题，可能没考虑更多的边界问题，但细节忘光了，工作比较忙，你可以尝试自己做实验，看看在实验上的效果，如果有兴趣可以帮我提个pr,感谢。我目前对这份代码只能保证这份code可以跑出readme上的效果。

@hangil55 i dont understand what your mean. The params in DeformableConvolution: bottom[0](data): (batch_size, channel, height, width) bottom[1] (offset): (batch_size, deformable_group * kernel[0] * kernel[1]*2, height, width) you can read the...

sorry, the codes dont support matlab

https://github.com/unsky/Deformable-ConvNets-caffe/issues/13

the setting of batch norm is very very import in detection task,you can refer the setting in my model. ps:i think there have mang bugs in your test

@RuiminChen https://github.com/unsky/Deformable-ConvNets-caffe/blob/master/deformable_conv_faster-rcnn/models/pascal_voc/ResNet50/faster_rcnn_end2end/train.prototxt i think we should fix scale params

sorry, i dont meet this problem, you should debug by yourself.

update readme

https://github.com/BVLC/caffe/issues/4324

修改这个是因为一些人反应无法加载yml文件，所以写在了config.py，正常情况下 yml会覆盖掉config里的一些参数，而且全部实用“gt”进行训练的，ss是fast rcnn遗留的一些东西，在这么没有用途