T2M

Reported on 64 benchmarks across 4 tasks

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Audio32 results

10-shot image generationonHumanML3D
Diversity
9.175
best: 23.692 (Diffuion Motion)
10-shot image generationonHumanML3D
FID
1.087
best: 0.028 (Motion Anything)
10-shot image generationonHumanML3D
Multimodality
2.219
best: 4.139 (TM2D (t2m))
10-shot image generationonHumanML3D
R Precision Top3
0.736
best: 0.842 (MotionPCM)
10-shot image generationonInter-X
FID
5.481
best: 0.399 (InterMask)
10-shot image generationonInter-X
MMDist
9.576
best: 9.58 (InterGen)
10-shot image generationonInter-X
MModality
2.761
best: 3.686 (InterGen)
10-shot image generationonInter-X
R-Precision Top3
0.396
best: 0.705 (InterMask)
10-shot image generationonInterHuman
FID
13.769
best: 5.154 (InterMask)
10-shot image generationonInterHuman
MMDist
5.731
best: 7.125 (MDM)
10-shot image generationonInterHuman
MModality
1.387
best: 2.35 (MDM)
10-shot image generationonInterHuman
R-Precision Top3
0.464
best: 0.687 (in2IN)
10-shot image generationonKIT Motion-Language
Diversity
10.72
best: 11.198 (T2M-GPT (τ = 0))
10-shot image generationonKIT Motion-Language
FID
3.022
best: 0.131 (Motion Anything)
10-shot image generationonKIT Motion-Language
Multimodality
2.052
best: 3.627 (GraphMotion)
10-shot image generationonKIT Motion-Language
R Precision Top3
0.681
best: 0.803 (BiPO)
3D Human Pose TrackingonHumanML3D
Diversity
9.175
best: 23.692 (Diffuion Motion)
3D Human Pose TrackingonHumanML3D
FID
1.087
best: 0.028 (Motion Anything)
3D Human Pose TrackingonHumanML3D
Multimodality
2.219
best: 4.139 (TM2D (t2m))
3D Human Pose TrackingonHumanML3D
R Precision Top3
0.736
best: 0.842 (MotionPCM)
3D Human Pose TrackingonInter-X
FID
5.481
best: 0.399 (InterMask)
3D Human Pose TrackingonInter-X
MMDist
9.576
best: 9.58 (InterGen)
3D Human Pose TrackingonInter-X
MModality
2.761
best: 3.686 (InterGen)
3D Human Pose TrackingonInter-X
R-Precision Top3
0.396
best: 0.705 (InterMask)
3D Human Pose TrackingonInterHuman
FID
13.769
best: 5.154 (InterMask)
3D Human Pose TrackingonInterHuman
MMDist
5.731
best: 7.125 (MDM)
3D Human Pose TrackingonInterHuman
MModality
1.387
best: 2.35 (MDM)
3D Human Pose TrackingonInterHuman
R-Precision Top3
0.464
best: 0.687 (in2IN)
3D Human Pose TrackingonKIT Motion-Language
Diversity
10.72
best: 11.198 (T2M-GPT (τ = 0))
3D Human Pose TrackingonKIT Motion-Language
FID
3.022
best: 0.131 (Motion Anything)
3D Human Pose TrackingonKIT Motion-Language
Multimodality
2.052
best: 3.627 (GraphMotion)
3D Human Pose TrackingonKIT Motion-Language
R Precision Top3
0.681
best: 0.803 (BiPO)

Computer Vision16 results

Pose TrackingonHumanML3D
Diversity
9.175
best: 23.692 (Diffuion Motion)
Pose TrackingonHumanML3D
FID
1.087
best: 0.028 (Motion Anything)
Pose TrackingonHumanML3D
Multimodality
2.219
best: 4.139 (TM2D (t2m))
Pose TrackingonHumanML3D
R Precision Top3
0.736
best: 0.842 (MotionPCM)
Pose TrackingonInter-X
FID
5.481
best: 0.399 (InterMask)
Pose TrackingonInter-X
MMDist
9.576
best: 9.58 (InterGen)
Pose TrackingonInter-X
MModality
2.761
best: 3.686 (InterGen)
Pose TrackingonInter-X
R-Precision Top3
0.396
best: 0.705 (InterMask)
Pose TrackingonInterHuman
FID
13.769
best: 5.154 (InterMask)
Pose TrackingonInterHuman
MMDist
5.731
best: 7.125 (MDM)
Pose TrackingonInterHuman
MModality
1.387
best: 2.35 (MDM)
Pose TrackingonInterHuman
R-Precision Top3
0.464
best: 0.687 (in2IN)
Pose TrackingonKIT Motion-Language
Diversity
10.72
best: 11.198 (T2M-GPT (τ = 0))
Pose TrackingonKIT Motion-Language
FID
3.022
best: 0.131 (Motion Anything)
Pose TrackingonKIT Motion-Language
Multimodality
2.052
best: 3.627 (GraphMotion)
Pose TrackingonKIT Motion-Language
R Precision Top3
0.681
best: 0.803 (BiPO)

Computer Code16 results

Motion SynthesisonHumanML3D
Diversity
9.175
best: 23.692 (Diffuion Motion)
Motion SynthesisonHumanML3D
FID
1.087
best: 0.028 (Motion Anything)
Motion SynthesisonHumanML3D
Multimodality
2.219
best: 4.139 (TM2D (t2m))
Motion SynthesisonHumanML3D
R Precision Top3
0.736
best: 0.842 (MotionPCM)
Motion SynthesisonInter-X
FID
5.481
best: 0.399 (InterMask)
Motion SynthesisonInter-X
MMDist
9.576
best: 9.58 (InterGen)
Motion SynthesisonInter-X
MModality
2.761
best: 3.686 (InterGen)
Motion SynthesisonInter-X
R-Precision Top3
0.396
best: 0.705 (InterMask)
Motion SynthesisonInterHuman
FID
13.769
best: 5.154 (InterMask)
Motion SynthesisonInterHuman
MMDist
5.731
best: 7.125 (MDM)
Motion SynthesisonInterHuman
MModality
1.387
best: 2.35 (MDM)
Motion SynthesisonInterHuman
R-Precision Top3
0.464
best: 0.687 (in2IN)
Motion SynthesisonKIT Motion-Language
Diversity
10.72
best: 11.198 (T2M-GPT (τ = 0))
Motion SynthesisonKIT Motion-Language
FID
3.022
best: 0.131 (Motion Anything)
Motion SynthesisonKIT Motion-Language
Multimodality
2.052
best: 3.627 (GraphMotion)
Motion SynthesisonKIT Motion-Language
R Precision Top3
0.681
best: 0.803 (BiPO)