# SuDA: Support-based Domain Adaptation for Sim2Real Motion Capture with Flexible Sensors

arXiv: https://arxiv.org/abs/2405.16152 · HTML: https://mlsanity.com/paper/2405.16152/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 60 references that are in the map.
- First posted on arXiv 25 May 2024.

## Builds on (most cited first, 26 of 26)

- [Unsupervised Domain Adaptation by Backpropagation](https://mlsanity.com/paper/1409.7495/) (2014)
- [Learning Transferable Features with Deep Adaptation Networks](https://mlsanity.com/paper/1502.02791/) (2015)
- [How transferable are features in deep neural networks?](https://mlsanity.com/paper/1411.1792/) (2014)
- [Deep CORAL: Correlation Alignment for Deep Domain Adaptation](https://mlsanity.com/paper/1607.01719/) (2016)
- [Deep Domain Confusion: Maximizing for Domain Invariance](https://mlsanity.com/paper/1412.3474/) (2014)
- [End-to-End Human Pose and Mesh Reconstruction with Transformers](https://mlsanity.com/paper/2012.09760/) (2020)
- [Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks](https://mlsanity.com/paper/1612.05424/) (2016)
- [Characterizing and Avoiding Negative Transfer](https://mlsanity.com/paper/1811.09751/) (2018)
- [Physical Inertial Poser (PIP): Physics-aware Real-time Human Motion Tracking from Sparse Inertial Sensors](https://mlsanity.com/paper/2203.08528/) (2022)
- [Recovering 3D Human Mesh from Monocular Images: A Survey](https://mlsanity.com/paper/2203.01923/) (2022)
- [Driving Policy Transfer via Modularity and Abstraction](https://mlsanity.com/paper/1804.09364/) (2018)
- [QuestSim: Human Motion Tracking from Sparse Sensors with Simulated Avatars](https://mlsanity.com/paper/2209.09391/) (2022)
- [Sim2real transfer learning for 3D human pose estimation: motion to the rescue](https://mlsanity.com/paper/1907.02499/) (2019)
- [MOTSynth: How Can Synthetic Data Help Pedestrian Detection and Tracking?](https://mlsanity.com/paper/2108.09518/) (2021)
- [X-Avatar: Expressive Human Avatars](https://mlsanity.com/paper/2303.04805/) (2023)
- [Does computer vision matter for action?](https://mlsanity.com/paper/1905.12887/) (2019)
- [IMUPoser: Full-Body Pose Estimation using IMUs in Phones, Watches, and Earbuds](https://mlsanity.com/paper/2304.12518/) (2023)
- [Transfer Learning with Dynamic Adversarial Adaptation Network](https://mlsanity.com/paper/1909.08184/) (2019)
- [Adversarial Training Helps Transfer Learning via Better Representations](https://mlsanity.com/paper/2106.10189/) (2021)
- [Balanced Distribution Adaptation for Transfer Learning](https://mlsanity.com/paper/1807.00516/) (2018)
- [i-Sim2Real: Reinforcement Learning of Robotic Policies in Tight Human-Robot Interaction Loops](https://mlsanity.com/paper/2207.06572/) (2022)
- [RLSbench: Domain Adaptation Under Relaxed Label Shift](https://mlsanity.com/paper/2302.03020/) (2023)
- [Object Detection Using Sim2Real Domain Randomization for Robotic Applications](https://mlsanity.com/paper/2208.04171/) (2022)
- [A Primer on Motion Capture with Deep Learning: Principles, Pitfalls and Perspectives](https://mlsanity.com/paper/2009.00564/) (2020)
- [Sim2real Transfer Learning for Point Cloud Segmentation: An Industrial Application Case on Autonomous Disassembly](https://mlsanity.com/paper/2301.05033/) (2023)
- [Bi-Manual Block Assembly via Sim-to-Real Reinforcement Learning](https://mlsanity.com/paper/2303.14870/) (2023)

## Abstract (arXiv)

Flexible sensors hold promise for human motion capture (MoCap), offering advantages such as wearability, privacy preservation, and minimal constraints on natural movement. However, existing flexible sensor-based MoCap methods rely on deep learning and necessitate large and diverse labeled datasets for training. These data typically need to be collected in MoCap studios with specialized equipment and substantial manual labor, making them difficult and expensive to obtain at scale. Thanks to the high-linearity of flexible sensors, we address this challenge by proposing a novel Sim2Real Mocap solution based on domain adaptation, eliminating the need for labeled data yet achieving comparable accuracy to supervised learning. Our solution relies on a novel Support-based Domain Adaptation method, namely SuDA, which aligns the supports of the predictive functions rather than the instance-dependent distributions between the source and target domains. Extensive experimental results demonstrate the effectiveness of our method andits superiority over state-of-the-art distribution-based domain adaptation methods in our task.