# SIGMAN:Scaling 3D Human Gaussian Generation with Millions of Assets

arXiv: https://arxiv.org/abs/2504.06982 · HTML: https://mlsanity.com/paper/2504.06982/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 73 of its 94 references that are in the map.
- First posted on arXiv 9 Apr 2025.

## Builds on (most cited first, 30 of 73)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer](https://mlsanity.com/paper/2408.06072/) (2024)
- [Expressive Body Capture: 3D Hands, Face, and Body from a Single Image](https://mlsanity.com/paper/1904.05866/) (2019)
- [Root Mean Square Layer Normalization](https://mlsanity.com/paper/1910.07467/) (2019)
- [Human Motion Diffusion Model](https://mlsanity.com/paper/2209.14916/) (2022)
- [Cosmos World Foundation Model Platform for Physical AI](https://mlsanity.com/paper/2501.03575/) (2025)
- [PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization](https://mlsanity.com/paper/1905.05172/) (2019)
- [Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling](https://mlsanity.com/paper/2501.17811/) (2025)
- [Structured 3D Latents for Scalable and Versatile 3D Generation](https://mlsanity.com/paper/2412.01506/) (2024)
- [LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation](https://mlsanity.com/paper/2402.05054/) (2024)
- [Animate Anyone: Consistent and Controllable Image-to-Video Synthesis for Character Animation](https://mlsanity.com/paper/2311.17117/) (2023)
- [T2M-GPT: Generating Human Motion from Textual Descriptions with Discrete Representations](https://mlsanity.com/paper/2301.06052/) (2023)
- [PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization](https://mlsanity.com/paper/2004.00452/) (2020)
- [CLAY: A Controllable Large-scale Generative Model for Creating High-quality 3D Assets](https://mlsanity.com/paper/2406.13897/) (2024)
- [Animatable Neural Radiance Fields for Modeling Dynamic Human Bodies](https://mlsanity.com/paper/2105.02872/) (2021)
- [Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering](https://mlsanity.com/paper/2106.02634/) (2021)
- [ICON: Implicit Clothed humans Obtained from Normals](https://mlsanity.com/paper/2112.09127/) (2021)
- [Function4D: Real-time Human Volumetric Capture from Very Sparse Consumer RGBD Sensors](https://mlsanity.com/paper/2105.01859/) (2021)
- [PaMIR: Parametric Model-Conditioned Implicit Representation for Image-based Human Reconstruction](https://mlsanity.com/paper/2007.03858/) (2020)
- [Sapiens: Foundation for Human Vision Models](https://mlsanity.com/paper/2408.12569/) (2024)

## Built on by (most cited first, 12 of 12)

- [Go to Zero: Towards Zero-shot Motion Generation with Million-scale Data](https://mlsanity.com/paper/2507.07095/) (2025)
- [SyncHuman: Synchronizing 2D and 3D Generative Models for Single-view Human Reconstruction](https://mlsanity.com/paper/2510.07723/) (2025)
- [LHM++: An Efficient Large Human Reconstruction Model for Pose-free Images to 3D](https://mlsanity.com/paper/2506.13766/) (2025)
- [Gloria: Consistent Character Video Generation via Content Anchors](https://mlsanity.com/paper/2603.29931/) (2026)
- [Snap-Snap: Taking Two Images to Reconstruct 3D Human Gaussians in Milliseconds](https://mlsanity.com/paper/2508.14892/) (2025)
- [HuGeDiff: 3D Human Generation via Diffusion with Gaussian Splatting](https://mlsanity.com/paper/2506.04351/) (2025)
- [End-to-End Spatial-Temporal Transformer for Real-time 4D HOI Reconstruction](https://mlsanity.com/paper/2603.14435/) (2026)
- [GenLCA: 3D Diffusion for Full-Body Avatars from In-the-Wild Videos](https://mlsanity.com/paper/2604.07273/) (2026)
- [Large-Scale High-Quality 3D Gaussian Head Reconstruction from Multi-View Captures](https://mlsanity.com/paper/2605.04035/) (2026)
- [LogiShot: Logically Coherent Cross-Shot Video Generation](https://mlsanity.com/paper/2608.08820/) (2026)
- [NoPo-Avatar: Generalizable and Animatable Avatars from Sparse Inputs without Human Poses](https://mlsanity.com/paper/2511.16673/) (2025)
- [HumanNOVA: Photorealistic, Universal and Rapid 3D Human Avatar Modeling from a Single Image](https://mlsanity.com/paper/2606.02573/) (2026)

## Abstract (arXiv)

3D human digitization has long been a highly pursued yet challenging task. Existing methods aim to generate high-quality 3D digital humans from single or multiple views, but remain primarily constrained by current paradigms and the scarcity of 3D human assets. Specifically, recent approaches fall into several paradigms: optimization-based and feed-forward (both single-view regression and multi-view generation with reconstruction). However, they are limited by slow speed, low quality, cascade reasoning, and ambiguity in mapping low-dimensional planes to high-dimensional space due to occlusion and invisibility, respectively. Furthermore, existing 3D human assets remain small-scale, insufficient for large-scale training. To address these challenges, we propose a latent space generation paradigm for 3D human digitization, which involves compressing multi-view images into Gaussians via a UV-structured VAE, along with DiT-based conditional generation, we transform the ill-posed low-to-high-dimensional mapping problem into a learnable distribution shift, which also supports end-to-end inference. In addition, we employ the multi-view optimization approach combined with synthetic data to construct the HGS-1M dataset, which contains 1 million 3D Gaussian assets to support the large-scale training. Experimental results demonstrate that our paradigm, powered by large-scale training, produces high-quality 3D human Gaussians with intricate textures, facial details, and loose clothing deformation.