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.