HRNet-Human-Pose-Estimation
HRNet-Human-Pose-Estimation copied to clipboard
HRNet
This repo is copied from https://github.com/leoxiaobin/deep-high-resolution-net.pytorch
Thank you for this repo. How did you modify the network for classification? For example, did you simply add a 1000d dense layer to your final high-res heatmap?
How well the model performs in occlusion scenario and tracking scenario?
The MPII dataset training can only achieve the accuracy of 89.8 on MPII VAL, which cannot reach the 90.3 in the paper
Now, i want to inference model hrnet with hrnet_w32_coco_256x192-c78dce93_20200708.pth checkpoint, i want to know information of output tensor. I used Netron.app, i know the output as under image 
我想问一下，coco_pose_hrnet_w32_256x192.pth这个权重是否训练模型时代表着输入的图片，高度为256，宽度为192？就像论文中说的一样： 
Hello, Thank you for your contribution. Why is there a 0.5 factor applied to the loss in JointMSELoss? I didn't find any mention of this in the paper. Also, why...
Hi! I would like to use **HRNet** for pose estimation with my **custom dataset**. I have cloned the repository, installed all the required packages from the requirements.txt file, and adjusted...
