This is a PyTorch implementation of Cascaded Refinement Network described in
Results on DAVIS2016 can be download here
- PyTorch
- Opencv
- Edit
img_path
,gt_path
,list_path
in the file./CRN_Pascal_Pretrain/config/CRN_Pascal.cfg
. - Edit file
./CRN_Pascal_Pretrain/train.sh
. - Run
sh train.sh
.
- Edit
img_path
,gt_path
,list_path
in the file./CRN_DAVIS16_Pretrain/config/CRN_DAVIS16.cfg
. - Copy a Pascal-pretrained model to
./CRN_DAVIS16_Pretrain/trained_model/
. - Edit file
./CRN_DAVIS16_Pretrain/train.sh
. - Run
sh train.sh
.
- Edit
img_path
,gt_path
,list_path
in the file./CRN_DAVIS16_Oneshot/config/CRN_DAVIS16.cfg
. - Copy a DAVIS16-pretrained model to
./CRN_DAVIS16_Oneshot/trained_model/
. - Edit file
./CRN_DAVIS16_Pretrain/train.sh
. - Run
sh train.sh
.
@InProceedings{Hu_2018_CVPR,
author = {Hu, Ping and Wang, Gang and Kong , Xiangfei and Kuen, Jason and Tan, Yap-Peng},
title = {Motion-Guided Cascaded Refinement Network for Video Object Segmentation},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
year = {2018}
}