-
I want to retrain FOMM with 68 keypoints and release it publicly, if successful. I haven't used Google Cloud yet or trained FOMM, but if it's feasible, I'm going to research more about it. I was wonde…
ghost updated
2 years ago
-
I'm currently using opencv 3.2.0. While I was compiling I got this error:
`fatal error: opencv2/xfeatures2d/nonfree.hpp: No such file or directory compilation terminated.`
even if I have correct…
-
In your current code:
- you define a rectangle on the forehead.
- Then use the 4 corners and do openCV optical flow keypoint tracking to get smoother tracking
- use estimateAffine2D to rectify th…
-
if imageSize equal 224， it's realtime，but keypoint‘s coordinate is incor。
how can i do to make the keypoint's coordiante coorect？thaks a lot！
-
Hi, I'm trying to merge the pose inference code with our own method， but I found that the inference model require the dxy which is calculated by predicted keypoints and reprojected keypoints as a par…
-
Hi,
When I run the gen_run.py to preprocess my own data. For the step 10, errors are shown below when running optimize_smpl.py file.
###########
can not find keypoints for 00000.png …
-
Hey there,
I am trying to add some new keypoints to the detection to generate some additionel PAF's.
To get something like the Body25 modell, from the "Open Pose" paper.
In the README it sad…
-
Dear Author,
I really like your work and have a few questions I'd like to ask you.
1. Firstly, if the keypoint tracking method is trained on a single task with 100 human demonstrations and 10 ro…
-
-
Great work! I am wondering how we can get an occlusion / visibility score of each keypoint for a given camera? Like can we determine whether this keypoint is occluded by other parts of hand or body?