https://github.com/xsxszab/dhsnet-pytorch
Pytorch implementation of DHSNet(CVPR2016)
https://github.com/xsxszab/dhsnet-pytorch
pytorch salient-object-detection
Last synced: 6 months ago
JSON representation
Pytorch implementation of DHSNet(CVPR2016)
- Host: GitHub
- URL: https://github.com/xsxszab/dhsnet-pytorch
- Owner: xsxszab
- Created: 2020-01-24T01:29:18.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2020-01-24T01:37:33.000Z (over 5 years ago)
- Last Synced: 2025-02-11T18:45:54.077Z (8 months ago)
- Topics: pytorch, salient-object-detection
- Language: Python
- Homepage:
- Size: 7.81 KB
- Stars: 4
- Watchers: 2
- Forks: 0
- Open Issues: 2
-
Metadata Files:
- Readme: readme.md
Awesome Lists containing this project
README
## DHS net
DHS net for salient objects detection(Pytorch implementation).
Some part of this project is based on
codes from https://github.com/wlguan/DHSNet-PyTorch ,
and this project is an optimized version.
### Requirements
Original running environment:
* Python 3.7.5
* Pytorch 1.3.1
* TorchVision 0.2.1
* pillow 7.0.0
See requirements.txt for detail.
### Training
1. Put corresponding dataset in ./input/
* training images(RGB, jpg format): ./input/train/raw/
* training masks(gray, png format): ./input/train/mask/
* validation images(RGB, jpg format): ./input/test/raw/
* validation masks(gray, png format): ./input/test/mask/
2. Run train.py, if you want to change some parameters,
see train.py for detail.
### Inference
1. Put inference data in ./inference/
* inference images(RGB, jpg format): ./inference
2. Run inference.py, output saliency maps will be in
./output directory.