Open momo1986 opened 5 years ago
Hi Sampepose,
Many thanks for your flownet2 model implementation.
Currently, I can use the pre-traiend model to generate the optical-flow image.
Is there any demo that we can use the flownet2 model and generated optical-flow images to recognize different actions, e.g., clip, waive or run.
Thanks a lot! Regards!
I have same question for using flownet2 in action recognition task. For example, UCF101 datasets
Hi Sampepose,
Many thanks for your flownet2 model implementation.
Currently, I can use the pre-traiend model to generate the optical-flow image.
Is there any demo that we can use the flownet2 model and generated optical-flow images to recognize different actions, e.g., clip, waive or run.
Thanks a lot! Regards!