Awesome
Caffe Implementation (v1) for
Learning Unsupervised Video Object Segmentation through Visual Attention (CVPR19)
Paying Attention to Video Object Pattern Understanding (PAMI20)
===========================================================================
:fire::fire::fire:Update
2019/10: Results on DAVIS17 val and test sets are added (instance-level video object segmentation)
===========================================================================
-
Please install our modified caffe first.
-
Download the model weight from https://drive.google.com/open?id=1aO2nAaQMy-A76NjRSQDTx53AfpqyWUCt and put dvap_agos.caffemodel into the 'model' folder.
-
Then edit paths in './test/test.py' and './test/davis_bs3.txt'.
-
After you get the original results from the network, you can use './test/davis_crf.py' to process them.
-
Our results on DAVIS16, FBMS and Youtube can be find at 'seg_results.rar'. Saliency results can be found at https://drive.google.com/file/d/1qtr2h69MEkBYLno_KqYIuiwZGFYmSHr4/view?usp=sharing
-
Human eye fixation data for DAVIS16, Youtube and Segtrackv2 can be found at https://drive.google.com/open?id=11LAg69lnCp7TfWFYjntPm4RK4KfgL8_Y.
-
Our model, segmentation and saliency results, and fixation data can also be found at Baidu Pan https://pan.baidu.com/s/1hz6iVOAMW3QRORebK3mC3Q , extaction code: 6prn
If you find our project helpful, your citations are highly appreciated:
@InProceedings{Wang_2019_CVPR,
author = {Wang, Wenguan and Song, Hongmei and Zhao, Shuyang and Shen, Jianbing and Zhao, Sanyuan and Hoi, Steven Chu Hong and Ling, Haibin},
title = {Learning Unsupervised Video Object Segmentation through Visual Attention},
booktitle = {CVPR},
year = {2019}
}
@InProceedings{Song_2018_ECCV,
author = {Song, Hongmei and Wang, Wenguan and Zhao, Sanyuan and Shen, Jianbing and Lam, Kin-Man},
title = {Pyramid Dilated Deeper ConvLSTM for Video Salient Object Detection},
booktitle = {ECCV},
year = {2018}
}
Other related projects/papers:
Zero-shot Video Object Segmentation via Attentive Graph Neural Networks (ICCV2019 Oral)
Saliency-Aware Geodesic Video Object Segmentation (CVPR15)
======================================================================== This repository is no longer maintained.