# MAGE:A Multi-stage Avatar Generator with Sparse Observations

arXiv: https://arxiv.org/abs/2505.06411 · HTML: https://mlsanity.com/paper/2505.06411/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 43 references that are in the map.
- First posted on arXiv 9 May 2025.

## Builds on (most cited first, 24 of 24)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Improved Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2102.09672/) (2021)
- [GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models](https://mlsanity.com/paper/2112.10741/) (2021)
- [Variational Inference with Normalizing Flows](https://mlsanity.com/paper/1505.05770/) (2015)
- [AMASS: Archive of Motion Capture as Surface Shapes](https://mlsanity.com/paper/1904.03278/) (2019)
- [On the Continuity of Rotation Representations in Neural Networks](https://mlsanity.com/paper/1812.07035/) (2018)
- [AI Choreographer: Music Conditioned 3D Dance Generation with AIST++](https://mlsanity.com/paper/2101.08779/) (2021)
- [TM2T: Stochastic and Tokenized Modeling for the Reciprocal Generation of 3D Human Motions and Texts](https://mlsanity.com/paper/2207.01696/) (2022)
- [Structural-RNN: Deep Learning on Spatio-Temporal Graphs](https://mlsanity.com/paper/1511.05298/) (2015)
- [Physical Inertial Poser (PIP): Physics-aware Real-time Human Motion Tracking from Sparse Inertial Sensors](https://mlsanity.com/paper/2203.08528/) (2022)
- [AvatarPoser: Articulated Full-Body Pose Tracking from Sparse Motion Sensing](https://mlsanity.com/paper/2207.13784/) (2022)
- [Sparse Inertial Poser: Automatic 3D Human Pose Estimation from Sparse IMUs](https://mlsanity.com/paper/1703.08014/) (2017)
- [Avatars Grow Legs: Generating Smooth Human Motion from Sparse Tracking Inputs with Diffusion Model](https://mlsanity.com/paper/2304.08577/) (2023)
- [Transformer Inertial Poser: Real-time Human Motion Reconstruction from Sparse IMUs with Simultaneous Terrain Generation](https://mlsanity.com/paper/2203.15720/) (2022)
- [FLAG: Flow-based 3D Avatar Generation from Sparse Observations](https://mlsanity.com/paper/2203.05789/) (2022)
- [Rhythm is a Dancer: Music-Driven Motion Synthesis with Global Structure](https://mlsanity.com/paper/2111.12159/) (2021)
- [LoBSTr: Real-time Lower-body Pose Prediction from Sparse Upper-body Tracking Signals](https://mlsanity.com/paper/2103.01500/) (2021)
- [BoDiffusion: Diffusing Sparse Observations for Full-Body Human Motion Synthesis](https://mlsanity.com/paper/2304.11118/) (2023)
- [Realistic Full-Body Tracking from Sparse Observations via Joint-Level Modeling](https://mlsanity.com/paper/2308.08855/) (2023)
- [GUESS:GradUally Enriching SyntheSis for Text-Driven Human Motion Generation](https://mlsanity.com/paper/2401.02142/) (2024)
- [Stratified Avatar Generation from Sparse Observations](https://mlsanity.com/paper/2405.20786/) (2024)

## Abstract (arXiv)

Inferring full-body poses from Head Mounted Devices, which capture only 3-joint observations from the head and wrists, is a challenging task with wide AR/VR applications. Previous attempts focus on learning one-stage motion mapping and thus suffer from an over-large inference space for unobserved body joint motions. This often leads to unsatisfactory lower-body predictions and poor temporal consistency, resulting in unrealistic or incoherent motion sequences. To address this, we propose a powerful Multi-stage Avatar GEnerator named MAGE that factorizes this one-stage direct motion mapping learning with a progressive prediction strategy. Specifically, given initial 3-joint motions, MAGE gradually inferring multi-scale body part poses at different abstract granularity levels, starting from a 6-part body representation and gradually refining to 22 joints. With decreasing abstract levels step by step, MAGE introduces more motion context priors from former prediction stages and thus improves realistic motion completion with richer constraint conditions and less ambiguity. Extensive experiments on large-scale datasets verify that MAGE significantly outperforms state-of-the-art methods with better accuracy and continuity.