paper :Bidirectional Feature Pyramid Network with Recurrent Attention Residual Modules for Shadow Detection
github:BDRAR
Preparation
- Set the path of pretrained ResNeXt model in resnext/config.py
- Set the path of SBU dataset in config.py
The pretrained ResNeXt model is ported from the official torch version, using the convertor provided by clcarwin. You can directly download the pretrained model ported by me.
效果展示:
以下左邊是true shadow mask,右圖是predict mask,下面是GT
good:
not good :
1.本身是類似陰影的黑色物體被誤判爲陰影,當然這可能我訓練的模型效果有關
2.漏檢部分:
整體來講,效果還是闊以的。