# 3D-Consistent Human Avatars with Sparse Inputs via Gaussian Splatting and Contrastive Learning

arXiv: https://arxiv.org/abs/2408.09663 · HTML: https://mlsanity.com/paper/2408.09663/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 33 of its 52 references that are in the map.
- First posted on arXiv 19 Aug 2024.

## Builds on (most cited first, 30 of 33)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://mlsanity.com/paper/1612.00593/) (2016)
- [Dynamic Graph CNN for Learning on Point Clouds](https://mlsanity.com/paper/1801.07829/) (2018)
- [Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervision](https://mlsanity.com/paper/1912.07372/) (2019)
- [UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstruction](https://mlsanity.com/paper/2104.10078/) (2021)
- [Neural Body: Implicit Neural Representations with Structured Latent Codes for Novel View Synthesis of Dynamic Humans](https://mlsanity.com/paper/2012.15838/) (2020)
- [SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering](https://mlsanity.com/paper/2311.12775/) (2023)
- [PointContrast: Unsupervised Pre-training for 3D Point Cloud Understanding](https://mlsanity.com/paper/2007.10985/) (2020)
- [HumanNeRF: Free-viewpoint Rendering of Moving People from Monocular Video](https://mlsanity.com/paper/2201.04127/) (2022)
- [PointCLIP: Point Cloud Understanding by CLIP](https://mlsanity.com/paper/2112.02413/) (2021)
- [SC-GS: Sparse-Controlled Gaussian Splatting for Editable Dynamic Scenes](https://mlsanity.com/paper/2312.14937/) (2023)
- [GaussianEditor: Swift and Controllable 3D Editing with Gaussian Splatting](https://mlsanity.com/paper/2311.14521/) (2023)
- [Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering](https://mlsanity.com/paper/2106.02634/) (2021)
- [Differentiable Surface Splatting for Point-based Geometry Processing](https://mlsanity.com/paper/1906.04173/) (2019)
- [Neural Articulated Radiance Field](https://mlsanity.com/paper/2104.03110/) (2021)
- [3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting](https://mlsanity.com/paper/2312.09228/) (2023)
- [GaussianAvatar: Towards Realistic Human Avatar Modeling from a Single Video via Animatable 3D Gaussians](https://mlsanity.com/paper/2312.02134/) (2023)
- [Shape As Points: A Differentiable Poisson Solver](https://mlsanity.com/paper/2106.03452/) (2021)
- [SplattingAvatar: Realistic Real-Time Human Avatars with Mesh-Embedded Gaussian Splatting](https://mlsanity.com/paper/2403.05087/) (2024)
- [Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining](https://mlsanity.com/paper/2302.02318/) (2023)
- [ARAH: Animatable Volume Rendering of Articulated Human SDFs](https://mlsanity.com/paper/2210.10036/) (2022)
- [LEAP: Learning Articulated Occupancy of People](https://mlsanity.com/paper/2104.06849/) (2021)
- [Neural Cages for Detail-Preserving 3D Deformations](https://mlsanity.com/paper/1912.06395/) (2019)
- [Human Gaussian Splatting: Real-time Rendering of Animatable Avatars](https://mlsanity.com/paper/2311.17113/) (2023)
- [MonoHuman: Animatable Human Neural Field from Monocular Video](https://mlsanity.com/paper/2304.02001/) (2023)
- [View-Consistent 3D Editing with Gaussian Splatting](https://mlsanity.com/paper/2403.11868/) (2024)
- [GoMAvatar: Efficient Animatable Human Modeling from Monocular Video Using Gaussians-on-Mesh](https://mlsanity.com/paper/2404.07991/) (2024)
- [PoseVocab: Learning Joint-structured Pose Embeddings for Human Avatar Modeling](https://mlsanity.com/paper/2304.13006/) (2023)
- [Geometry-Guided Progressive NeRF for Generalizable and Efficient Neural Human Rendering](https://mlsanity.com/paper/2112.04312/) (2021)

## Built on by (most cited first, 6 of 6)

- [Efficient Physics Simulation for 3D Scenes via MLLM-Guided Gaussian Splatting](https://mlsanity.com/paper/2411.12789/) (2024)
- [The NeRF Signature: Codebook-Aided Watermarking for Neural Radiance Fields](https://mlsanity.com/paper/2502.19125/) (2025)
- [MoGA: 3D Generative Avatar Prior for Monocular Gaussian Avatar Reconstruction](https://mlsanity.com/paper/2507.23597/) (2025)
- [Towards Synthesized and Editable Motion In-Betweening Through Part-Wise Phase Representation](https://mlsanity.com/paper/2503.08180/) (2025)
- [Deblur-Avatar: Animatable Avatars from Motion-Blurred Monocular Videos](https://mlsanity.com/paper/2501.13335/) (2025)
- [TeleOpBench: A Simulator-Centric Benchmark for Dual-Arm Dexterous Teleoperation](https://mlsanity.com/paper/2505.12748/) (2025)

## Abstract (arXiv)

Existing approaches for human avatar generation--both NeRF-based and 3D Gaussian Splatting (3DGS) based--struggle with maintaining 3D consistency and exhibit degraded detail reconstruction, particularly when training with sparse inputs. To address this challenge, we propose CHASE, a novel framework that achieves dense-input-level performance using only sparse inputs through two key innovations: cross-pose intrinsic 3D consistency supervision and 3D geometry contrastive learning. Building upon prior skeleton-driven approaches that combine rigid deformation with non-rigid cloth dynamics, we first establish baseline avatars with fundamental 3D consistency. To enhance 3D consistency under sparse inputs, we introduce a Dynamic Avatar Adjustment (DAA) module, which refines deformed Gaussians by leveraging similar poses from the training set. By minimizing the rendering discrepancy between adjusted Gaussians and reference poses, DAA provides additional supervision for avatar reconstruction. We further maintain global 3D consistency through a novel geometry-aware contrastive learning strategy. While designed for sparse inputs, CHASE surpasses state-of-the-art methods across both full and sparse settings on ZJU-MoCap and H36M datasets, demonstrating that our enhanced 3D consistency leads to superior rendering quality.