Closed t-qureshi closed 3 years ago
Hi, thank you for bringing this up. flownet.pt is the pertrained global flow field net. You can obtain it either from training by script/run_pose.sh
(and specify --flownet_path $THE_PATH) or extract the weights of global flow field net from GFLA.
However, if you only wants to run inferences, you shouldn't have needed it, as it would be replaced by "latest_net_Flow.pt" in the checkpoints. I updated the code configuration, so I think it will no longer be a problem. If you don't want to update the newer version of code, try specify --flownet_path ''
which should bypass the default path of flownet
Hi, Thank You so much for sharing this amazing work. I was thinking to implement your work but I couldn't find flownet.pt. Can you please share the model Thanks