pytorch_Realtime_Multi-Person_Pose_Estimation
pytorch_Realtime_Multi-Person_Pose_Estimation copied to clipboard
tensorboy
hello authors, i have trained a simple model using FCN for face keypoint regression. Can you please tell me how we can get heatmaps of muliple faces in the bottom-up...
In the demo code, update the lines to load the model as below: model = get_model('vgg19') wts_dict = torch.load(args.weight) wts_dict_corrected = {k.replace('module.',''): v for k, v in wts_dict.items()} model.load_state_dict(wts_dict_corrected) _Originally...
I train the model from scratch, but I got the results below which is very bad. @tensorboy .Could you tell me some reason? I did not change any code. 
Epoch: [0][0/565] Data time 6.441 (6.441) Loss 0.4377 (0.4377)loss_stage1_L1: 0.1094 (0.1094) loss_stage1_L2: 0.0547 (0.0547) loss_stage2_L1: 0.0038 (0.0038) loss_stage2_L2: 0.0512 (0.0512) loss_stage3_L1: 0.0035 (0.0035) loss_stage3_L2: 0.0512 (0.0512) loss_stage4_L1: 0.0035 (0.0035) loss_stage4_L2:...
(sort2) hxy@uac516:~/Desktop/anomaly_detection/pytorch_Realtime_Multi-Person_Pose_Estimation$ python demo/picture_demo.py Bulding VGG19 Traceback (most recent call last): File "demo/picture_demo.py", line 46, in model.load_state_dict(torch.load(args.weight)) File "/home/hxy/anaconda3/envs/sort2/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1224, in load_state_dict self.__class__.__name__, "\n\t".join(error_msgs))) RuntimeError: Error(s) in loading state_dict...
Hi, Thanks for sharing this great repo. One thing I'd like to know is how to train my own dataset. Can you provide more code so we can train from...
Could you please share other links to download the pre-trained model? cause the dropbbox seems to be unconnected.
Just curious, from the model you converted directly do you get the same results as the original openpose COCO model?
In lib/network/rtpose_shufflenetV2.py, from network import slim from network.slim import g_name but I can not find where is "slim"
