Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Human Motion Editing on HumanML3D (test)
Loading...
0
MPJPE
Ground-Truth
-0.042
0.2415
0.525
0.8085
Dec 6, 2024
MPJPE
FID
Updated 4d ago
Evaluation Results
Method
Method
Links
MPJPE
FID
Ground-Truth
2024.12
0
0
CigTime
Mode=Full-parameter tu...
2024.12
0.13
1.44
Llama-3-8B
Mode=In-context learning
2024.12
0.21
3.04
Mistral-7B
Mode=In-context learning
2024.12
0.22
5.03
Llama-3-8B-LoRA
Mode=LoRA fine-tuning
2024.12
0.24
2.09
Mistral-7B-LoRA
Mode=LoRA fine-tuning
2024.12
0.75
1.84
MotionGPT
Input Template=Section...
2024.12
0.8
8.84
MotionGPT-M2T
Description=Motion cap...
2024.12
1.05
7.96
Feedback
Search any
task
Search any
task