MoMask: Generative Masked Modeling of 3D Human Motions
Authors:
Preprint
Abstract
We introduce MoMask, a novel masked modeling framework for text-driven 3D human motion generation. In MoMask, a hierarchical quantization scheme is employed to represent human motion as multi-layer discrete motion tokens with high-fidelity details. Starting at the base layer, with a sequence of motion tokens obtained by vector quantization, the residual tokens of increasing orders are derived and stored at the subsequent layers of the hierarchy. This is consequently followed by two distinct bidirectional transformers. For the base-layer motion tokens, a Masked Transformer is designated to predict randomly masked motion tokens conditioned on text input at training stage. During generation (i.e. inference) stage, starting from an empty sequence, our Masked Transformer iteratively fills up the missing tokens; Subsequently, a Residual Transformer learns to progressively predict the next-layer tokens based on the results from current layer. Extensive experiments demonstrate that MoMask outperforms the state-of-art methods on the text-to-motion generation task, with an FID of 0.045 (vs e.g. 0.141 of T2M-GPT) on the HumanML3D dataset, and 0.228 (vs 0.514) on KIT-ML, respectively. MoMask can also be seamlessly applied in related tasks without further model fine-tuning, such as text-guided temporal inpainting.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- Neural Codec Language Models are Zero-Shot Text to Speech SynthesizersIEEE Transactions on Audio, Speech and Language Processing, 2025
- TM2D: Bimodality Driven 3D Dance Generation via Music-Text Integration2023
- ReMoDiffuse: Retrieval-Augmented Motion Diffusion Model2023
- AudioLM: A Language Modeling Approach to Audio GenerationIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2023
- Generating Human Motion from Textual Descriptions with Discrete Representations2023
- Investigating Pose Representations and Motion Contexts Modeling for 3D Motion PredictionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2022
- Action2video: Generating Videos of Human 3D ActionsInternational Journal of Computer Vision, 2022
- SoundStream: An End-to-End Neural Audio CodecIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021
- Learning a bidirectional mapping between human whole-body motion and natural language using deep recurrent neural networksRobotics and Autonomous Systems, 2018
- The KIT Motion-Language DatasetBig Data, 2016