# VideoMDM: Towards 3D Human Motion Generation From 2D Supervision

arXiv: https://arxiv.org/abs/2606.13364 · HTML: https://mlsanity.com/paper/2606.13364/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 46 of its 55 references that are in the map.
- First posted on arXiv 11 Jun 2026.

## Builds on (most cited first, 30 of 46)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [DreamFusion: Text-to-3D using 2D Diffusion](https://mlsanity.com/paper/2209.14988/) (2022)
- [Zero-1-to-3: Zero-shot One Image to 3D Object](https://mlsanity.com/paper/2303.11328/) (2023)
- [Deep High-Resolution Representation Learning for Visual Recognition](https://mlsanity.com/paper/1908.07919/) (2019)
- [Demystifying MMD GANs](https://mlsanity.com/paper/1801.01401/) (2018)
- [AMASS: Archive of Motion Capture as Surface Shapes](https://mlsanity.com/paper/1904.03278/) (2019)
- [ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillation](https://mlsanity.com/paper/2305.16213/) (2023)
- [OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields](https://mlsanity.com/paper/1812.08008/) (2018)
- [Human Motion Diffusion Model](https://mlsanity.com/paper/2209.14916/) (2022)
- [Improved Precision and Recall Metric for Assessing Generative Models](https://mlsanity.com/paper/1904.06991/) (2019)
- [MotionDiffuse: Text-Driven Human Motion Generation with Diffusion Model](https://mlsanity.com/paper/2208.15001/) (2022)
- [Objaverse-XL: A Universe of 10M+ 3D Objects](https://mlsanity.com/paper/2307.05663/) (2023)
- [Executing your Commands via Motion Diffusion in Latent Space](https://mlsanity.com/paper/2212.04048/) (2022)
- [One-2-3-45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape Optimization](https://mlsanity.com/paper/2306.16928/) (2023)
- [LION: Latent Point Diffusion Models for 3D Shape Generation](https://mlsanity.com/paper/2210.06978/) (2022)
- [Zero123++: a Single Image to Consistent Multi-view Diffusion Base Model](https://mlsanity.com/paper/2310.15110/) (2023)
- [Action2Motion: Conditioned Generation of 3D Human Motions](https://mlsanity.com/paper/2007.15240/) (2020)
- [CAT3D: Create Anything in 3D with Multi-View Diffusion Models](https://mlsanity.com/paper/2405.10314/) (2024)
- [MoMask: Generative Masked Modeling of 3D Human Motions](https://mlsanity.com/paper/2312.00063/) (2023)
- [One-2-3-45++: Fast Single Image to 3D Objects with Consistent Multi-View Generation and 3D Diffusion](https://mlsanity.com/paper/2311.07885/) (2023)
- [Guided Motion Diffusion for Controllable Human Motion Synthesis](https://mlsanity.com/paper/2305.12577/) (2023)
- [WHAM: Reconstructing World-grounded Humans with Accurate 3D Motion](https://mlsanity.com/paper/2312.07531/) (2023)
- [AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Time](https://mlsanity.com/paper/2211.03375/) (2022)
- [MeshDiffusion: Score-based Generative 3D Mesh Modeling](https://mlsanity.com/paper/2303.08133/) (2023)
- [MotionBERT: A Unified Perspective on Learning Human Motion Representations](https://mlsanity.com/paper/2210.06551/) (2022)

## Abstract (arXiv)

We introduce VideoMDM, a diffusion-based framework that trains 3D human motion priors directly from accurate 2D poses extracted from monocular videos, without any 3D ground truth. A pretrained 2D-to-3D lifter provides approximate 3D pose sequences that serve as a noisy teacher: these are diffused, denoised by the model in 3D, and supervised in 2D by reprojecting the prediction and comparing against accurate keypoints. We show that, under mild assumptions, a depth-weighted 2D reprojection loss is equivalent in expectation to direct 3D supervision, and we adapt standard 3D motion regularizers - velocity consistency and over-parameterized representation alignment - to this 2D setting. Unlike methods that lift 2D to 3D only at inference, VideoMDM learns a coherent 3D motion manifold during training. On HumanML3D it nearly closes the gap to fully 3D-supervised MDM (FID 0.88 vs 0.54); On real video datasets Fit3D and NBA the method learns to generate motions consistently preferred by humans, with strong quantitative results.