deep-motion-editing icon indicating copy to clipboard operation
deep-motion-editing copied to clipboard

An end-to-end library for editing and rendering motion of 3D characters with deep learning [SIGGRAPH 2020]

Results 64 deep-motion-editing issues
Sort by recently updated
recently updated
newest added

感谢分享您的工作! 在使用重定向的工程中,我使用了lafan1的骨架作为输入,但当运行demo.py后,出现如下模型加载的问题,请问这种情况我是需要使用该骨架重新训练吗? Traceback (most recent call last): File "eval_single_pair.py", line 97, in main() File "eval_single_pair.py", line 76, in main model.load(epoch=20000) File "E:\python\motion_editing\retargeting\models\architecture.py", line 274, in load model.load(os.path.join(self.model_save_dir, 'topology{}'.format(i)), epoch) File...

李老师,您好, 非常兴奋看到这么棒的开源工作,实际用下来,有以下几个问题想要请教您一下,还望不吝赐教 1、如何用咱们的demo.py脚本把一个bvh动画retargeting到一个无动画的bvh角色上? 如果我有一个无动画的fbx静态网格体,已经按照fbx2bvh处理成一个静态的bvh文件,preprocess.py是无法获取mean和var的,并且.npy文件也是null,那我如何把一个动画retargeting到一个静态的bvh文件上呢?而不是示例中的,两个都是带有动画的bvh文件,才能执行demo.py 2、如果是两个都带有动画的bvh文件,进行retargeting,是不是必须要保证两个的动画长度一致?我发现动画不一样,也会出错,如下: ![cXK3ZAHgC8](https://github.com/DeepMotionEditing/deep-motion-editing/assets/31311358/589ceda1-2588-4599-b839-bffd2825ab2c) 3、咱们论文中有提到homeomorphic skeleton,我理解只要是正常的人形骨架,比如metahuman的骨骼、ue小白人的骨骼,应该都满足和miaxmo骨骼的homeomorphic吧? 4、如果2成立的话,咱们的预训练模型,是不是可以直接在miaxmo骨骼、metahuman骨骼、ue小白人骨骼之间进行cross_structure的retargeting 谢谢~ 期待李老师您的回复

李老师: 您好,感谢你们在重定向方面开源了这么棒的工作,受益匪浅。我在你们的数据集上实验均没有问题,但是使用自己的数据集bvh进行训练时发现效果不佳,我用的数据集是Trinity Speech-Gesture dataset,在blender中打开具有这样的特点:正常动作面朝的方向与Tpose相反,如下图 ![image](https://user-images.githubusercontent.com/105349375/195482379-6a5badeb-6ee2-44f3-8f5b-bb25a06d06b2.png) ![image](https://user-images.githubusercontent.com/105349375/195482411-e01b9fa5-f9c9-4a9f-bca0-4d229bc22f0e.png) 针对该数据集新定义了简化后的骨骼 corps_name_trinity = ['Hips', "LeftUpLeg", "LeftLeg", "LeftFoot", "LeftForeFoot", "LeftToeBase", "RightUpLeg", "RightLeg", "RightFoot", "RightForeFoot", "RightToeBase", 'Spine', 'Spine1', 'Spine2', "Spine3", 'Neck', 'Neck1', 'Head', 'LeftShoulder', 'LeftArm', 'LeftForeArm',...

I've the following understanding: The idea is to see if the Generator, when trained exclusively on the small batch, can produce samples that the Discriminator thinks are real. If your...

hello there, thanks alot for this awesome work, i need help on how to retarget a bvh file to an acceptable bone format to be able to use it in...

Q1, the src.bvh and dst.bvh must be the same Skeletal hierarchy ? or main body skeletons are the same Skeletal hierarchy ? if i use smpl_24, kinectv2_25, h36m_17, h36m_25, mpi_inf_3dhp_17,...

When I run the following command line, the following error occurred. ``` $ python3 train.py --save_dir=./training/ load from file ./datasets/Mixamo/Aj.npy Window count: 179, total frame (without downsampling): 21437 load from...

When I run the followint command line, the following error occurred. ``` $ python3 datasets/preprocess.py Traceback (most recent call last): File "datasets/preprocess.py", line 4, in from datasets.bvh_parser import BVH_file ModuleNotFoundError:...