video prediction

[JWFanggit]

Hello, your work is very attractive. Can this method be used as a video frame prediction task?

[zhangjiewu]

thank you @JWFanggit for the comment. i believe that extending spatial attention to spatiotemporal, which is one of the core ideas of our method, could benefit general temporal modeling. so yes, i think it can be adapted to other video-related tasks, such as video prediction, completion, interpolation, among others. i look forward to seeing the community's progress in this direction.

[JWFanggit]

Thank you very much for your reply！