Skip to content

Latest commit

 

History

History
41 lines (32 loc) · 1.26 KB

README.md

File metadata and controls

41 lines (32 loc) · 1.26 KB
Raw

DQSD(TIP 2020)

DQSD：Depth Quality Aware Salient Object Detection(Under review)
The network structure is as follows:

Requirment

  • tensorflow1.4
  • python3.6

Preparation

1.Clone code by clone git https://github.com/qdu1995/DQSD.git
2.Download the official pre training weight Resnet or download initial model link(jsf8)
3.Download rgb-d dataset datasets in the folder of data for training or test.

for train:

1.Put the train dataset into the train folder.
2.Put the pretrained model into the 'model' folder.
3.python train.py
datalink(cdky)

for test:

1.Download testing data, put the test dataset into the test folder
2.Download pretrained model link(0vo7), put the model into the 'model' folder
3.python demo.py

Result:

data link(j4iz)

Evalution:

Download the evaluation indicators link(vhfc).Thanks for the authors (http://dpfan.net/)

If you think this work is helpful, please cite: @article{

}
qdu1995@163.com