/ViSalientObject

Video Salient Object Detection via Fully Convolutional Networks (TIP18)

Primary LanguagePython

Keras Implementation (v1) for

Video salient object detection via fully convolutional networks, IEEE Trans. on Image Processing, 27(1):38-49, 2018

By Wenguan Wang and Jianbing Shen and Ling Shao

========================================================================

Please install keras first

The model weights and results on DAVIS and FBMS datasets can be downloaded from

Baidu Wangpan:https://pan.baidu.com/s/1dE22MFR

password: jmph

or from google drive: https://drive.google.com/drive/u/0/folders/1mc6nnr8RrMZwAXjV0XS9Hv7QJE4quvkx

Please put the models under the folder 'videosalientobjectdetection' and run main.py.

Our results (MAE) are sightly improved over the original scores reported in our paper:

DAVIS: 0.0587 FBMS: 0.0619

You can find our results on UVSD dataset here: https://pan.baidu.com/s/1DUWHiIf-EGG1l-izaAf09Q

Our code for synthesizing video data from still image has been uploaded.

UnZip synthesizevideo.rar first.

Then install vlfeat toolbox first (in 'functions' folder).

Run synthesizevideo.m.

========================================================================

If you find our method useful in your research, please consider citing the following papers:

  1. W. Wang, J. Shen, and L. Shao, Video salient object detection via fully convolutional networks, IEEE Trans. on Image Processing, 27(1):38-49, 2018

  2. W. Wang, J. Shen, and L. Shao, Consistent video saliency using local gradient flow optimization and global refinement, IEEE Trans. on Image Processing, 24(11):4185-4196, 2015

  3. W. Wang, J. Shen, R. Yang, and F. Porikli, Saliency-aware video object segmentation, IEEE Trans. on Pattern Analysis and Machine Intelligence, 40(1):20-33, 2018

  4. W. Wang, J. Shen, F. Porikli, Saliency-aware geodesic video object segmentation, IEEE CVPR, pp. 3395-3402, 2015

========================================================================

Other related projects:

(ECCV18)Pyramid Dilated Deeper ConvLSTM for Video Salient Object Detection

(CVPR19) Learning Unsupervised Video Object Segmentation through Visual Attention

(CVPR19 Oral) Shifting More Attention to Video Salient Object Detection

(CVPR19)See More, Know More: Unsupervised Video Object Segmentation With Co-Attention Siamese Networks

========================================================================

Any comments, please email: wenguanwang.china@gmail.com