pytorch_Realtime_Multi-Person_Pose_Estimation
pytorch_Realtime_Multi-Person_Pose_Estimation copied to clipboard
tensorboy
Hi , I am getting the below error. Please help me . RuntimeError: Error(s) in loading state_dict for rtpose_model: Missing key(s) in state_dict: "model0.0.bias", "model0.0.weight", "model0.2.bias", "model0.2.weight", "model0.5.bias", "model0.5.weight", "model0.7.bias",...
hi, @tensorboy ,when I run the web_demo, I met the following error: "Bulding VGG19 Traceback (most recent call last): File "web_demo.py", line 92, in multiplier, oriImg, model, 'rtpose') TypeError: get_outputs()...
Hi, What kind of fps do you get while testing? For me it takes about 5.2 seconds per frame on 1080 Ti, for video resolution of 1280x720. Are other people...
Thank you for your great work! I tried to train the model with shufflenet, but the loss turn to Nan. Can you give me some advices? ``` Epoch: [0][0/1520] Data...
Thanks for the great repo! It helped me alot to understand open-pose. I have tried training with train_vgg.py, to train the model from scratch and the loss seemed to have...
@tensorboy when I training this model use the train_VGG19.py, and I found that the GPU's memory has doubled, do you know why? Look forward to your reply! thanks
Hello, I can't get close results with the "trained from scratch model" with train_VGG.py. And in the training process, the validation loss is always 0.0, and I haven't found the...
Hi, during inference, i get the following error: " RuntimeError('cuda runtime error (2) : out of memory at /pytorch/torch/lib/THC/THCTensorCopy.cu:204',), )" Which usually happens when using very large sized images. Even...
https://github.com/tensorboy/pytorch_Realtime_Multi-Person_Pose_Estimation/blob/5ce5a4fbeb3cd6ffcd7aec18d1b5779d518a8736/evaluate/coco_eval.py#L153 https://github.com/tensorboy/pytorch_Realtime_Multi-Person_Pose_Estimation/blob/5ce5a4fbeb3cd6ffcd7aec18d1b5779d518a8736/evaluate/coco_eval.py#L159
I am looking for python eval code for rtpose, i find many issues on rtpose ,maybe 0.577 should be the result of rtpose of first 2644 images of val2014? and...
