deep-motion-editing

deep-motion-editing copied to clipboard

eval_prepare looks dirty + some other questions

5

https://github.com/DeepMotionEditing/deep-motion-editing/blob/master/retargeting/eval_single_pair.py#L8 I just don't get why character names are hardcoded here. Also don't understand the meaning of the "_m" suffix of some characters. Are i'm right that it has an...

TheCrazyT

Retargeting question

5

I use intra-structure retargeting. I change the "input.bvh", while the "gt.bvh" remains the original one. But the "result.bvh" style is almost the "gt.bvh" but not "input.bvh". Is it right? I'm...

chaserat

Can I just train the intra type without the cross type？

2

excuse me,Can I just train the intra type without the cross type? @kfiraberman

wuxiaolianggit

[Feature Enhancement] Motion retargeting from OpenPose JSON output

4

First of all, thanks for all the work. Currently we can use style transfer from JSON output [as given in README.md](https://github.com/DeepMotionEditing/deep-motion-editing#style-from-videos) Can we use the JSON output from OpenPose, for...

aitikgupta

Do I need to train new models for new characters in retargeting?

3

Hi, all I read your papers carefully, you mentioned that "since the domains have different dimensions, the two networks (A → B, A → B) cannot share weights, so they...

DetectiveLYJ

How to test with customer data?

2

Hi, When I use other characters on the Mixamo dataset in retarget work, I will need a std file, but I didn't find the code to calculate the std. Could...

DetectiveLYJ

非标准躯干数据集的可能性

1

李老师您好！ 我想咨询一下就是我现在有一份不标准的躯干的bvh数据，这是因为我选取了一套挥杆动作，高尔夫球杆绑定在人的左手上，也就是左手会比右手多出一些关节，这样不对称的躯干会对训练产生影响吗，我自己尝试训练了一下并尝试了intra的重定向但是效果很不理想，是因为末梢点变成了球杆的末梢导致的吗，烦请李老师百忙之余回答一下，感谢！  

dddyyyplus

Retargeting retraining on customized dataset

3

Hello Peizhuo, Thanks for making this amazing project public available! I am trying to retrain the retargeting model on my own dataset, I have a new skeleton and its motion...

HaozhouPang

处理过手指的重定向吗？

3

您好，请问有使用此方法处理过手指吗？ 我用自己的数据训练了模型，重定向后的身体动作基本一致，但是手指部分还是存在一些问题，以下用其中一个动作来示意说明，原动作如下图所示：  重定向后的手指如下图所示：  主要存在的问题是手指关节的角度有较明显的差别，详细如下： 1、左手四指指端没有伸直并拢，左手大拇指没有弯曲并拢； 2、右手四指弯曲的角度不对。 我大致的思路是，在bvh_parser.py对骨架扩展了所有的手指关节，手指的末端作为末端关节，去掉gan后进行训练，训练的epoch到2k，loss曲线如下：  。

anddyhzw

How the the skining work?

1

Hi, from the skining script I can tell anywhere saved the skined fbx. Doesn't it for applying bvh animation to target fbx and generated new fbx with animation? I can...

lucasjinreal