Remove the rain in the input image/video by applying the derain methods based on convolutional neural networks. Supported models:
Recurrent Squeeze-and-Excitation Context Aggregation Net (RESCAN). See http://openaccess.thecvf.com/content_ECCV_2018/papers/Xia_Li_Recurrent_Squeeze-and-Excitation_Context_ECCV_2018_paper.pdf.
Training scripts as well as scripts for model generation are provided in the repository at https://github.com/XueweiMeng/derain_filter.git.
The filter accepts the following options:
‘dnn_backend’
Specify which DNN backend to use for model loading and execution. This option accepts the following values:
‘native’
Native implementation of DNN loading and execution.
Default value is ‘native’.
‘model’
Set path to model file specifying network architecture and its parameters. Note that different backends use different file formats. TensorFlow backend can load files for both formats, while native backend can load files for only its format.
ffmpeg -i derain_input.mp4 -vf derain=model=derain_RESCAN.model derain_output.mp4
缺点：利用 CPU 处理时速度太慢，效果不是很完美。
 |
 |
