Skip to content

Latest commit

 

History

History
27 lines (21 loc) · 987 Bytes

README.md

File metadata and controls

27 lines (21 loc) · 987 Bytes
Raw

Segment removal using the FCN-ResNet101

The remove-bg.py file is a python script that removes all the parts of an image except the regions that represent a class that we want to detect. It then saves the resulting picture with name 'removed_bg.jpg'

Packages Required

  • numpy
  • torch
  • torchvision
  • PIL
  • matplotlib

Instructions

to run the program:

git clone https://github.com/ChuckTG/Segment_Removal_Pytorch_FCN-ResNet.git
cd Segment_Removal_Pytorch_FCN-ResNet

python remove_bg.py

Example

This is an output produced by the script:

References

The FCN-ResNet is loaded from torchvision.models.segmentation, it is a Fully Convolutional Network as proposed by Evan Shelhamer et.al and has the ResNet-101 model as backbone.