Keras implementation of PSPNet(caffe)

Implemented Architecture of pyramid scene parsing network in Keras

Converted trained weights needed to run the network.

Download converted weights here: link:pspnet50_ade20k.npy

And place in directory with pspnet50_ade20k.npy

Weights from caffemodel were converted by, weight_converter.py. The usage of this file is

python weight_converter.py <path to .prototxt> <path to .caffemodel>

Running this need to compile the original PSPNet caffe code and pycaffe.

Interpolation layer is implemented in code as custom layer "Interp"

Important

Results Keras:

Pycaffe result

Dependencies:

1. Tensorflow
2. Keras
3. numpy
4. pycaffe(PSPNet)(optional)

Usage:

python pspnet.py --input-path INPUT_PATH --output-path OUTPUT_PATH