- Linux
- Python 3.5+ (Say goodbye to Python2)
- PyTorch 1.0+ or PyTorch-nightly
- CUDA 9.0+
- NCCL 2+
- GCC 4.9+
- mmcv
We have tested the following versions of OS and softwares:
- OS: Ubuntu 16.04/18.04 and CentOS 7.2
- CUDA: 9.0/9.2/10.0
- NCCL: 2.1.15/2.2.13/2.3.7/2.4.2
- GCC: 4.9/5.3/5.4/7.3
-
Create a conda virtual environment and activate it. Then install Cython.
-
Install PyTorch stable or nightly and torchvision following the official instructions.
-
Clone the mmdetection repository.
git clone https://github.com/360-SSOD/download.git
- RPN models are available in the Model
360-SOD&360-SSOD Tencent Cloud Link
360-SSOD : Tencent Cloud Link
-
Please download our pretrained model.
Put this model in./model/
and./reimg/
. -
Run:
test.py
Saliency Evaluation Method : The code can be found in Ming-Ming Cheng and Deng-Ping Fan.
@article{ma2020stage,
title={Stage-wise salient object detection in 360° omnidirectional image via object-level semantical saliency ranking},
author={Ma, Guangxiao and Li, Shuai and Chen, Chenglizhao and Hao, Aimin and Qin, Hong},
journal={IEEE Transactions on Visualization and Computer Graphics},
volume={26},
number={12},
pages={3535--3545},
year={2020},
publisher={IEEE}
}