Yu Sun comments

Results 301 comments of


                                            Yu Sun

批处理视频用哪个文件

https://github.com/Arthur151/ROMP/tree/master/simple_romp 输入图片/视频，输出3D人体关键点/网格及空间位置信息。详情建议看一下论文，写的很简洁明了。

写个python脚本来执行。类似于这样： ``` import os import glob for video_path in glob.glob('/path/to/videos/*'): save_path = video_path.replace('.mp4','_results.mp4') os.system('romp --mode=video --calc_smpl --render_mesh --input={} --save_path={} --save_video'.format(video_path, save_path)) ```

unstable 3D translation (especially along depth) for monocular pose estimation

Hi, Zhengdi Yu, long time no see. 1. About the depth ordering during rendering, maybe this is more clear: https://github.com/Arthur151/ROMP/blob/e4613fd564cd632ac531228b94f949eaed76345e/simple_romp/vis_human/main.py#L40 2. Yes, I think BEV is more stable in depth....

unstable 3D translation (especially along depth) for monocular pose estimation

Yes, due to the usage of sim3dr, I think that directly convert cam to trans3d is of enough accuracy for sim3dr rendering usage. Therefore, I use it directly. But BEV...

训练自己的数据集

是的，可以参考[这个介绍](https://github.com/Arthur151/ROMP/blob/master/docs/train.md)里的Training with your own datasets

训练自己的数据集

这个暂时不太行。如果想要输出自定义的关节的旋转角，我知道的可能的方案： 1）需要制作一个类似于SMPL的统计形状模型。 2）将SMPL的姿态retargeting到具有这26个关节角的三维人体模型上。

Any ideas about to driving characters in Unreal Engine?

The pipeline of driving the model in Unreal Engine is supposed to be similar to that of Blender. I am not familiar with the interface of Unreal Engine to achieve...

Distributed training raise an error

Thanks for the bug report. Please replace it as romp.train like this https://github.com/Arthur151/ROMP/blob/master/scripts/train_distributed.sh

Distributed training raise an error

Oh, the command you use is different from what is on my rep. Besides, please make sure that you run the code under ROMP folder. ``` CUDA_VISIBLE_DEVICES=${GPUS} nohup python -u...