https://github.com/blade6570/dilation-pytorch-semantic-segmentation
A PyTorch implementation of semantic segmentation according to Multi-Scale Context Aggregation by Dilated Convolutions by Yu and Koltun.
https://github.com/blade6570/dilation-pytorch-semantic-segmentation
cityscapes cnn deep-learning python3 pytorch semantic-segmentation
Last synced: 5 months ago
JSON representation
A PyTorch implementation of semantic segmentation according to Multi-Scale Context Aggregation by Dilated Convolutions by Yu and Koltun.
- Host: GitHub
- URL: https://github.com/blade6570/dilation-pytorch-semantic-segmentation
- Owner: Blade6570
- Created: 2018-01-07T21:51:40.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2018-06-18T12:55:55.000Z (over 7 years ago)
- Last Synced: 2025-05-12T23:44:40.082Z (5 months ago)
- Topics: cityscapes, cnn, deep-learning, python3, pytorch, semantic-segmentation
- Language: Python
- Homepage:
- Size: 2.24 MB
- Stars: 36
- Watchers: 1
- Forks: 10
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Dilation-Pytorch-Semantic Segmentation
A PyTorch implementation of semantic segmentation according to [Multi-Scale Context Aggregation by Dilated Convolutions](https://arxiv.org/abs/1511.07122) by Yu and Koltun.
Pretrained weights are obtained from [dialation-Tensorflow](https://github.com/ndrplz/dilation-tensorflow) and thanks to [Andrea Palazzi](https://github.com/ndrplz) for his work from which this work has been inspired.
**How To**
1. Download the pretrained weights from [cityscapes](https://drive.google.com/file/d/0Bx9YaGcDPu3XR0d4cXVSWmtVdEE/view)
2. Keep that in the *data* folder
3. Run *CAN_network.py* for the demo.
**Notes**
I have kept the output image size 1024x1024x3 due to memory constraint. If you want the size to be 1024x2048x3, have enough memory and change the sizes appropriately in the code.