# Active Learning of Neural Collision Handler for Complex 3D Mesh Deformations

arXiv: https://arxiv.org/abs/2110.07727 · HTML: https://mlsanity.com/paper/2110.07727/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 51 references that are in the map.
- First posted on arXiv 8 Oct 2021.

## Builds on (most cited first, 23 of 23)

- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [Deep Bayesian Active Learning with Image Data](https://mlsanity.com/paper/1703.02910/) (2017)
- [AtlasNet: A Papier-Mach'e Approach to Learning 3D Surface Generation](https://mlsanity.com/paper/1802.05384/) (2018)
- [Interaction Networks for Learning about Objects, Relations and Physics](https://mlsanity.com/paper/1612.00222/) (2016)
- [Capture, Learning, and Synthesis of 3D Speaking Styles](https://mlsanity.com/paper/1905.03079/) (2019)
- [Hand Keypoint Detection in Single Images using Multiview Bootstrapping](https://mlsanity.com/paper/1704.07809/) (2017)
- [STAR: Sparse Trained Articulated Human Body Regressor](https://mlsanity.com/paper/2008.08535/) (2020)
- [TailorNet: Predicting Clothing in 3D as a Function of Human Pose, Shape and Garment Style](https://mlsanity.com/paper/2003.04583/) (2020)
- [Fast and Flexible Indoor Scene Synthesis via Deep Convolutional Generative Models](https://mlsanity.com/paper/1811.12463/) (2018)
- [DeepWrinkles: Accurate and Realistic Clothing Modeling](https://mlsanity.com/paper/1808.03417/) (2018)
- [On Self-Contact and Human Pose](https://mlsanity.com/paper/2104.03176/) (2021)
- [Deep Active Learning over the Long Tail](https://mlsanity.com/paper/1711.00941/) (2017)
- [Variational Autoencoders for Deforming 3D Mesh Models](https://mlsanity.com/paper/1709.04307/) (2017)
- [GarNet: A Two-Stream Network for Fast and Accurate 3D Cloth Draping](https://mlsanity.com/paper/1811.10983/) (2018)
- [Active Learning for Deep Detection Neural Networks](https://mlsanity.com/paper/1911.09168/) (2019)
- [PartNet: A Recursive Part Decomposition Network for Fine-grained and Hierarchical Shape Segmentation](https://mlsanity.com/paper/1903.00709/) (2019)
- [Learning Complex 3D Human Self-Contact](https://mlsanity.com/paper/2012.10366/) (2020)
- [Sparse Data Driven Mesh Deformation](https://mlsanity.com/paper/1709.01250/) (2017)
- [A Deep Emulator for Secondary Motion of 3D Characters](https://mlsanity.com/paper/2103.01261/) (2021)
- [LCollision: Fast Generation of Collision-Free Human Poses using Learned Non-Penetration Constraints](https://mlsanity.com/paper/2011.03632/) (2020)
- [Mesh-based Autoencoders for Localized Deformation Component Analysis](https://mlsanity.com/paper/1709.04304/) (2017)
- [Multiscale Mesh Deformation Component Analysis with Attention-based Autoencoders](https://mlsanity.com/paper/2012.02459/) (2020)
- [Fastron: An Online Learning-Based Model and Active Learning Strategy for Proxy Collision Detection](https://mlsanity.com/paper/1709.02316/) (2017)

## Built on by (most cited first, 5 of 5)

- [Human Trajectory Prediction via Neural Social Physics](https://mlsanity.com/paper/2207.10435/) (2022)
- [N-Cloth: Predicting 3D Cloth Deformation with Mesh-Based Networks](https://mlsanity.com/paper/2112.06397/) (2021)
- [A Repulsive Force Unit for Garment Collision Handling in Neural Networks](https://mlsanity.com/paper/2207.13871/) (2022)
- [CTSN: Predicting Cloth Deformation for Skeleton-based Characters with a Two-stream Skinning Network](https://mlsanity.com/paper/2305.18808/) (2023)
- [PoseShield: Neural Collision Fields for Human Self-Collision Resolution](https://mlsanity.com/paper/2606.29686/) (2026)

## Abstract (arXiv)

We present a robust learning algorithm to detect and handle collisions in 3D deforming meshes. Our collision detector is represented as a bilevel deep autoencoder with an attention mechanism that identifies colliding mesh sub-parts. We use a numerical optimization algorithm to resolve penetrations guided by the network. Our learned collision handler can resolve collisions for unseen, high-dimensional meshes with thousands of vertices. To obtain stable network performance in such large and unseen spaces, we progressively insert new collision data based on the errors in network inferences. We automatically label these data using an analytical collision detector and progressively fine-tune our detection networks. We evaluate our method for collision handling of complex, 3D meshes coming from several datasets with different shapes and topologies, including datasets corresponding to dressed and undressed human poses, cloth simulations, and human hand poses acquired using multiview capture systems. Our approach outperforms supervised learning methods and achieves 93.8-98.1% accuracy compared to the groundtruth by analytic methods. Compared to prior learning methods, our approach results in a 5.16%-25.50% lower false negative rate in terms of collision checking and a 9.65%-58.91% higher success rate in collision handling.