It is a implementation based on the following paper and the repo for the 2018 ICME Salient 360 challenge:
-
IEEE ICME Salient360! Grand Challenge: Visual attention modeling for 360 Content - 2018 edition.
https://salient360.ls2n.fr -
Kao Zhang, Zhenzhong Chen. Video Saliency Prediction Based on Spatial-Temporal Two-Stream Network. IEEE Trans. Circuits Syst. Video Techn. 2018.
Github: https://github.com/zhangkao/IIP_TwoS_Saliency
-
My test environment:
windows10, python3.6, keras2.1, tensorflow1.6, cuda9.0 -
My device:
GTX1080, 8G memory, Intel E5-CPU 32G RAM
How to run our code for task 1 and 2 ?
-
please change the working directory: "wkdir" to your path in the "zk_config.py" file, like
"wkdir = 'E:/Salient360-2018/Task1_2'" ; -
set the parameter "task_type = 'H'" or "task_type = 'HE'" for these two tasks in "zk_config.py" file;
-
set the parameter "with_CB = True"(default) or "with_CB = False" to control whether to use the center bias method or not;
-
put the test stimuli to "DataSet/Images/Stimuli" and "DataSet/Videos/Stimuli" folders in the "wkdir" path;
-
then run the demo "Test_images_Demo.py" and "Test_videos_Demo.py".
- The results of image task is saved by ".bin"(float32) and ".png" formats.
- The results of video task is saved by ".bin"(float32), ".mat"(int32) and ".mp4" formats.
- And it is easy to change the output format in our code.
If you use the video saliency model, please cite the following paper:
@article{Zhang2018Video,
author = {Kao Zhang and Zhenzhong Chen},
title = {Video Saliency Prediction Based on Spatial-Temporal Two-Stream Network},
journal = {IEEE Transactions on Circuits and Systems for Video Technology },
year = {2018}
}
Kao ZHANG
Laboratory of Intelligent Information Processing (LabIIP)
Wuhan University, Wuhan, China.
Email: zhangkao@whu.edu.cn
Zhenzhong CHEN (Professor and Director)
Laboratory of Intelligent Information Processing (LabIIP)
Wuhan University, Wuhan, China.
Email: zzchen@whu.edu.cn
Web: http://iip.whu.edu.cn/~zzchen/