mmpose

mmpose copied to clipboard

**Motivation**: We plan to improve our documentation and tutorials! Users may encounter various problems during the use of MMPose, and we plan to collect all these feedback and provide a...

jin-s13

No inference results on custom dataset

14

Hi, I have followed **macaque dataset** format to label the keypoints in custom dataset. My custom dataset comprises of 200 black and white .png images with straight lines. I have...

rubeea

innerlee

Meaning of the last two elements of `preds` in bottom up result

5

I try to print the array {'preds': [array([[9.70417969e+02, 4.94542969e+02, 1.00000000e+00, 1.00000000e+00, 1.00000000e+00] I only know the first x,y, keypoint_score . The other doesnt know. https://user-images.githubusercontent.com/59388989/117801610-1b1ac380-b287-11eb-990c-e37de7c3f39d.png

alicera

Add tldr-styled section in doc What is tldr: 

innerlee

Preprocess of human36m keypoint training data

12

Hello. What should I do if I want to train human36m 3d keypoint model? I only find the document with human36m mesh detection. And I can only find the procedures...

FireMonkey796

Add all-in-one graphic with all methods

3

There are a lot of methods this is difficult to compare, so I have the following suggestion: Add graphical picture for each data set with all methods with two axis:...

ioctl-user

[Feature Request] Add namespace when registering DistributedDataParallelWrapper

1

mmpose and mmgen both haveDistributedDataParallelWrapper. It would be better to register the hook in a namespace to avoid key-value conflict. related issue: https://github.com/open-mmlab/mmgeneration/issues/32

RangiLyu

android deployment

1

If I want to deploy mmpose application deploy in andoid mobile photo, how to do it.

wm901115nwpu

How much system ram is required per gpu for interhand3d dataset?

6

Looking at the [log provided](https://download.openmmlab.com/mmpose/hand3d/internet/res50_intehand3d_all_256x256_20210506.log.json) it looks like 8 Titan x gpus were used to train the interhand dataset with a batch size of 16 and 2 workers per gpu....

pablovela5620