fbx results not so good

[visonpon]

Hi, thanks for your generous sharing of this wonderful work, and I have tried to test on my own video.
when only look the mesh that projected onto the image, the results look good, but when converting to fbx and see in blender, although the pose looks normal, the root motion has big problems like drifting and jitter.

It seems this root motion problem are common in methods based on smpl, and I have noticed that the MTC and CHD have given some methods to improve the performance. but they are time-consuming.

I wonder if there may be a way to do some work combine these two works or you have some other ideas, thanks~

[Arthur151]

Great idea. I have been working on this problem too. Following the way that CHD explored, I am trying to develop a more concise/compact method to achieve the dynamics smoothness, just like the spirit of ROMP.
Welcome to have some dicussions about this.

[Arthur151]

@visonpon ,Hi,
I am trying to connect ROMP and contact_human_dynamics.
Could you please provide the output file of contact_human_dynamics, after running:
python run_totalcap.py --data ../data/example_data --out ../output/mtc_viz_out --totalcap ../external/MonocularTotalCapture

Thanks in advance

[visonpon]

Sorry for the delayed response, you can download from here, if you need other results from chd ,please let me know.

[Arthur151]

Thanks for your help. We really appreciate it.

[rlleshi]

First, thanks a lot for the great repo.

Do you have a rough timeline for when (if at all) this will be available?

[Arthur151]

@rlleshi Hi, thanks.
Sorry, I don't understand. When "what" will be available?

[rlleshi]

Following the way that CHD explored, I am trying to develop a more concise/compact method to achieve the dynamics smoothness, just like the spirit of ROMP.

I mean this.

[Arthur151]

I am still running for this goal. I expect that we can achieve this before the end of the year.