Hi, thank you for sharing the STRM code and great work on the paper!
I'm currently working on action understanding from static images and would like to use STRM for extracting action-related features. I noticed that the repository does not include pretrained model weights.
Could you please share:
Pretrained weights and expected input format, or any instructions for inference on custom data (like single images)?
This would be incredibly helpful for research use cases where training from scratch is not feasible.
Thanks in advance!
Hi, thank you for sharing the STRM code and great work on the paper!
I'm currently working on action understanding from static images and would like to use STRM for extracting action-related features. I noticed that the repository does not include pretrained model weights.
Could you please share:
Pretrained weights and expected input format, or any instructions for inference on custom data (like single images)?
This would be incredibly helpful for research use cases where training from scratch is not feasible.
Thanks in advance!