Official implementation of 'Learning Motion-Appearance Co-Attention for Zero-Shot Video Object Segmentation', in Pytorch.
'Learning Motion-Appearance Co-Attention for Zero-Shot Video Object Segmentation' has been accepted to ICCV 2021, as poster.
Code will come soon.
I am sorry about the delay of my code. Now i am doing research as an intern in Meituan and prepare for my AAAI Rebuttal and CVPR. Since I can't go back to school for the time being, I can't open source the code. If you need the model code, please email me. 2021-11-02