deep-high-resolution-net.pytorch

deep-high-resolution-net.pytorch copied to clipboard

How is the weight of each key point calculated

1

self.joints_weight = np.array( [ 1., 1., 1., 1., 1., 1., 1., 1.2, 1.2, 1.5, 1.5, 1., 1., 1.2, 1.2, 1.5, 1.5 ], dtype=np.float32 ).reshape((self.num_joints, 1)) Can you explain to me...

XiaopinWang

MPII Vis

2

wangdong0556

[Network structure] Why hrnet use basicblock in stage 2/3/4

3

Thanks for your works! In Deep residual network, resnet34 use basicblock and resnet50/101 use bottleneck for practice performance. So what about hrnet? If hrnet replace basicblock by bottleneck in stage...

HeathHose

RuntimeError: storage has wrong size: expected 0 got 128

1

Hi, I wanted to try using the pre-trained models with other data set. I am using azure ml to execute the code. I tried using the code: > python tools/test.py...

keyboard-user

Bumps [opencv-python](https://github.com/skvark/opencv-python) from 3.4.1.15 to 4.2.0.32. Release notes Sourced from opencv-python's releases. 4.2.0.32 opencv-python: https://pypi.org/project/opencv-python/ opencv-contrib-python: https://pypi.org/project/opencv-contrib-python/ opencv-python-headless: https://pypi.org/project/opencv-python-headless/ opencv-contrib-python-headless: https://pypi.org/project/opencv-contrib-python-headless/ OpenCV version 4.2.0. Changes: macOS environment updated from xcode8.3...

dependabot[bot]

How to install setup.py?

1

muxiddin19

ImportError: failed to map segment from shared object

1

Hello, can anyone help me to fix this error? I have made new fresh conda env and install pytorch 1.5.1 with cuda 10.2 and follow every single step from requirements.txt...

riorafe

AI challenger data not accessible? Corresponding weights is not provided. How to reproduce your result with extra data?

1

The AI challenger data is not accessible (https://challenger.ai/dataset/keypoint) The model weights trained with it is not provided? How to reproduce the best performed one of pose_hrnet_w48*?

ShuangjunLiu

why need norm = np.ones((pred.shape[0], 2)) * np.array([h, w]) / 10 in accuracy

1

in UDP-Pose/deep-high-resolution-net.pytorch/lib/core/evaluate.py Line 55 in d742edd norm = np.ones((pred.shape[0], 2)) * np.array([h, w]) / 10 norm = np.ones((pred.shape[0], 2)) * np.array([h, w]) / 10 I did know why use norm...

azuryl

Question about a smart detail

1

In lib/core/evaluate.py, function **calc_dists**, the Euclidean distance will be calculated under the condition: `if target[n, c, 0] > 1 and target[n, c, 1] > 1:` It seems that you exclude...

seekFire