deep-head-pose

deep-head-pose copied to clipboard

Hello @natanielruiz I have a few doubts that I would like to clarify. 1. Is this model suitable for estimating head pose during low-lit and night vision conditions? If yes,...

AjayBhat29

where is the filename list

2

I have download the 300W_LP set, but there is no file name list. Need I generate this list ? Thank you.

moonhyw

draw_axis

1

Hi team, Can someone provide me with the theory for the draw_axis function used to visualize the yaw, pitch, roll angle? Thanks

smonefrmsouth

Also, few changes that make it compatible with python 3

prakhargoyal106

what MSE actually does?

1

part of loss is mse ( predicted * idx_tensor, yaw ) ``` idx_tensor = [idx for idx in xrange(66)] idx_tensor = Variable(torch.FloatTensor(idx_tensor)).cuda(gpu) yaw_predicted = torch.sum(yaw_predicted * idx_tensor, 1) loss_reg_yaw =...

iperov

torch.autograd.backward(loss_seq, grad_seq),I got RuntimeError

3

Hi, I run the train_hopenet.py with my dataset ,pytorch version is 0.4.1, and got this error: File "XXX/train_hopenet.py", line 276, in `torch.autograd.backward(loss_seq, grad_seq)` File "XXX/torch/autograd/__init__.py",line 90, in backward `allow_unreachable=True) #...

cnnmena

I find in 300LP dataset, the flip sample label headpose doesn't match the origin sample label. For example the headpose of HELEN/HELEN_212515672_1_3.mat is (0.077, 0.27, -0.34) but HELEN_Flip/HELEN_212515672_1_3.mat is (0.019,...

vdgs

SushantGautam

do you have special training strategies? error:Yaw: 8.7961, Pitch: 6.7705, Roll: 5.8057, my training models can not reach your results as paper said

3

below is my best results. resnet_multi-loss: Test error in degrees of the model on the 1969 test images. Yaw: 8.7961, Pitch: 6.7705, Roll: 5.8057 resnet50 only: Test error in degrees...

maliho0803

how to generate the datasets of 300W_LP?

3

I have downloaded the datasets 300W_LP. I have a question about dealing with the datasets for training

mumuyanyan