-
Hello @chingswy,
Since you are updating the "Detector function" for yolo-hrnet, I tried to generate smpl parameters directly from 3d-keypoints of H36m.
However, I am unsure, if only 3d keypoints…
-
hello @chingswy ,
**So I asked this question in NeuralBody repo and Pengsida informed me to ask you:**
**Q1.** In **neural-body**, we did not plot the skeleton nor the key points as it is usuall…
-
(easymocap) E:\MOCAP\EasyMocap>python apps/demo/mv1p.py 0_input/project --out 1_output/project --vis_det --vis_repro --undis --sub_vis 1 2 3 4 --vis_smpl --end 650
Demo code for multiple views an…
-
Hi,
I was doing research on recomposing 3D model from OpenPose data using a single, uncalibrated, camera feed. I stumbled on your repo by chance (looking for stuff around the body25 model) and notice…
-
When making the code, I want to use my poses that I created myself, but online it's mostly 18 point bodies. I want the output like the article, what code should I use. thank you
-
Hi,
I am trying to use PySKL 3D with Body25 configuration (the one from OpenPose), is it possible to do it ?
-
(easymocap) M:\GITHUB\EASYMOCAP>python apps/demo/1v1p_mirror.py zju-m-demo/ --out zju-m-demo-output/ --vis_smpl --video
Demo code for single view and one person with mirror:
- Input : zju-…
-
The file `J_regressor_body25_smplx.txt` you provided is $118\times 10475$, while openpose output $137$ keypoints. How can I regress the total $137$ keypoints from SMPLX vertices? Could you please prov…
-
您好：
看了之前的issue，发现您提到这个文件是从别的项目引进来的但具体出处不可考了；麻烦您能介绍一下这个regressor是怎么生成的吗？
非常感谢！
ddlau updated
2 years ago
-
Hello there, I am reading your code about fitting SMPL pose parameters from 3D BODY25 joints.
I noticed that you have defined 5 types of losses (LossKeypoints3D, LossSmoothBodyMean, LossSmoothPoses,…