Light-T2M generates 3D human motion from text with 4.48M parameters, reporting FID 0.040 on HumanML3D (vs 0.045 for MoMask) and faster inference.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Light-T2M: A Lightweight and Fast Model for Text-to-motion Generation
Light-T2M generates 3D human motion from text with 4.48M parameters, reporting FID 0.040 on HumanML3D (vs 0.045 for MoMask) and faster inference.