# 3D Shape Variational Autoencoder Latent Disentanglement via Mini-Batch Feature Swapping for Bodies and Faces

arXiv: https://arxiv.org/abs/2111.12448 · HTML: https://mlsanity.com/paper/2111.12448/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 47 references that are in the map.
- First posted on arXiv 24 Nov 2021.

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

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [Dynamic Graph CNN for Learning on Point Clouds](https://mlsanity.com/paper/1801.07829/) (2018)
- [Disentangling by Factorising](https://mlsanity.com/paper/1802.05983/) (2018)
- [PointFlow: 3D Point Cloud Generation with Continuous Normalizing Flows](https://mlsanity.com/paper/1906.12320/) (2019)
- [Learning from Synthetic Humans](https://mlsanity.com/paper/1701.01370/) (2017)
- [3D Menagerie: Modeling the 3D shape and pose of animals](https://mlsanity.com/paper/1611.07700/) (2016)
- [3D Morphable Face Models -- Past, Present and Future](https://mlsanity.com/paper/1909.01815/) (2019)
- [Generating 3D faces using Convolutional Mesh Autoencoders](https://mlsanity.com/paper/1807.10267/) (2018)
- [Variational Inference of Disentangled Latent Concepts from Unlabeled Observations](https://mlsanity.com/paper/1711.00848/) (2017)
- [STAR: Sparse Trained Articulated Human Body Regressor](https://mlsanity.com/paper/2008.08535/) (2020)
- [Learning to Regress 3D Face Shape and Expression from an Image without 3D Supervision](https://mlsanity.com/paper/1905.06817/) (2019)
- [Deep Convolutional Inverse Graphics Network](https://mlsanity.com/paper/1503.03167/) (2015)
- [Learning Formation of Physically-Based Face Attributes](https://mlsanity.com/paper/2004.03458/) (2020)
- [SpiralNet++: A Fast and Highly Efficient Mesh Convolution Operator](https://mlsanity.com/paper/1911.05856/) (2019)
- [Variational Autoencoders Pursue PCA Directions (by Accident)](https://mlsanity.com/paper/1812.06775/) (2018)
- [Measuring Disentanglement: A Review of Metrics](https://mlsanity.com/paper/2012.09276/) (2020)
- [Towards a complete 3D morphable model of the human head](https://mlsanity.com/paper/1911.08008/) (2019)
- [Structured Disentangled Representations](https://mlsanity.com/paper/1804.02086/) (2018)
- [Guided Variational Autoencoder for Disentanglement Learning](https://mlsanity.com/paper/2004.01255/) (2020)
- [LIMP: Learning Latent Shape Representations with Metric Preservation Priors](https://mlsanity.com/paper/2003.12283/) (2020)
- [Fully Convolutional Mesh Autoencoder using Efficient Spatially Varying Kernels](https://mlsanity.com/paper/2006.04325/) (2020)
- [Deformable Shape Completion with Graph Convolutional Autoencoders](https://mlsanity.com/paper/1712.00268/) (2017)
- [Combining 3D Morphable Models: A Large scale Face-and-Head Model](https://mlsanity.com/paper/1903.03785/) (2019)
- [Unsupervised Model Selection for Variational Disentangled Representation Learning](https://mlsanity.com/paper/1905.12614/) (2019)
- [MeshGAN: Non-linear 3D Morphable Models of Faces](https://mlsanity.com/paper/1903.10384/) (2019)
- [Synthesizing Coupled 3D Face Modalities by Trunk-Branch Generative Adversarial Networks](https://mlsanity.com/paper/1909.02215/) (2019)
- [Geometric Disentanglement for Generative Latent Shape Models](https://mlsanity.com/paper/1908.06386/) (2019)
- [Mesh Variational Autoencoders with Edge Contraction Pooling](https://mlsanity.com/paper/1908.02507/) (2019)
- [Intraoperative Liver Surface Completion with Graph Convolutional VAE](https://mlsanity.com/paper/2009.03871/) (2020)

## Built on by (most cited first, 15 of 15)

- [Taming Video Models for 3D and 4D Generation via Zero-Shot Camera Control](https://mlsanity.com/paper/2509.15130/) (2025)
- [TUVF: Learning Generalizable Texture UV Radiance Fields](https://mlsanity.com/paper/2305.03040/) (2023)
- [ShapeFusion: A 3D diffusion model for localized shape editing](https://mlsanity.com/paper/2403.19773/) (2024)
- [3D Generative Model Latent Disentanglement via Local Eigenprojection](https://mlsanity.com/paper/2302.12798/) (2023)
- [Locally Adaptive Neural 3D Morphable Models](https://mlsanity.com/paper/2401.02937/) (2024)
- [SMPL-GPTexture: Dual-View 3D Human Texture Estimation using Text-to-Image Generation Models](https://mlsanity.com/paper/2504.13378/) (2025)
- [BFSM: 3D Bidirectional Face-Skull Morphable Model](https://mlsanity.com/paper/2509.24577/) (2025)
- [Parallelised Differentiable Straightest Geodesics for 3D Meshes](https://mlsanity.com/paper/2603.15780/) (2026)
- [Unsupervised Representation Learning for Diverse Deformable Shape Collections](https://mlsanity.com/paper/2310.18141/) (2023)
- [Disentangling Hippocampal Shape Variations: A Study of Neurological Disorders Using Mesh Variational Autoencoder with Contrastive Learning](https://mlsanity.com/paper/2404.00785/) (2024)
- [Representing 3D Faces with Learnable B-Spline Volumes](https://mlsanity.com/paper/2604.12894/) (2026)
- [Latent Disentanglement in Mesh Variational Autoencoders Improves the Diagnosis of Craniofacial Syndromes and Aids Surgical Planning](https://mlsanity.com/paper/2309.10825/) (2023)
- [MUNCH: Modelling Unique 'N Controllable Heads](https://mlsanity.com/paper/2310.02753/) (2023)
- [Plug-and-Play Feature Generation for Few-Shot Medical Image Classification](https://mlsanity.com/paper/2310.09471/) (2023)
- [Conditional Diffusion Model with Anatomical-Dose Dual Constraints for End-to-End Multi-Tumor Dose Prediction](https://mlsanity.com/paper/2508.02043/) (2025)

## Abstract (arXiv)

Learning a disentangled, interpretable, and structured latent representation in 3D generative models of faces and bodies is still an open problem. The problem is particularly acute when control over identity features is required. In this paper, we propose an intuitive yet effective self-supervised approach to train a 3D shape variational autoencoder (VAE) which encourages a disentangled latent representation of identity features. Curating the mini-batch generation by swapping arbitrary features across different shapes allows to define a loss function leveraging known differences and similarities in the latent representations. Experimental results conducted on 3D meshes show that state-of-the-art methods for latent disentanglement are not able to disentangle identity features of faces and bodies. Our proposed method properly decouples the generation of such features while maintaining good representation and reconstruction capabilities.