OpenGVLab/VideoMamba

About different Scan Methods Implementation

Opened this issue · 0 comments

Thank you for your open source excellent work.

I would like to ask about Fig. 4: Different scan methods.

From the code, it seems that only Spatial-First Bidirectional is supported.

Could you please indicate how the others (i.e., Temporal-First and Spatiotemporal versions) are implemented respectively?

Thank you, looking forward to your reply.