Deep-learning999

Hope to have Kinetics-TPS FineAction MultiSports data set support, pre-training model, training and web video inference demo

I hope to use posec3d to realize bone-based spatiotemporal motion detection

Request: I hope to provide pre-training of different frame lengths in slowonly and slowfast of ava-Kinetics, as well as pre-training of TimeSformer in ava2.1 and ava-Kinetics. I think the real-time...

Hope to have Kinetics-TPS FineAction MultiSports data set pre-training model

hrnet is very cumbersome, is there a lightweight model that can be applied to this algorithm

Does it need a vpn over the wall to work?

Only this information when running for 3 hours 0 -1 1 3520 models.common.Focus [3, 32, 3] 1 -1 1 18560 models.common.Conv [32, 64, 3, 2] 2 -1 1 18816 models.common.C3...

Epoch gpu_mem box obj cls kpt kptv total labels img_size 29/299 18G 0.0375 0.02693 0 0.2554 0.01235 0.3322 Class Images Labels P R [email protected] Traceback (most recent call last): File...