# Online Test-time Adaptation for 3D Human Pose Estimation: A Practical Perspective with Estimated 2D Poses

arXiv: https://arxiv.org/abs/2503.11194 · HTML: https://mlsanity.com/paper/2503.11194/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 36 references that are in the map.
- First posted on arXiv 14 Mar 2025.

## Builds on (most cited first, 27 of 27)

- [Deep High-Resolution Representation Learning for Human Pose Estimation](https://mlsanity.com/paper/1902.09212/) (2019)
- [Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields](https://mlsanity.com/paper/1611.08050/) (2016)
- [Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labels](https://mlsanity.com/paper/1804.06872/) (2018)
- [End-to-end Recovery of Human Shape and Pose](https://mlsanity.com/paper/1712.06584/) (2017)
- [Keep it SMPL: Automatic Estimation of 3D Human Pose and Shape from a Single Image](https://mlsanity.com/paper/1607.08128/) (2016)
- [Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loop](https://mlsanity.com/paper/1909.12828/) (2019)
- [Continual Test-Time Domain Adaptation](https://mlsanity.com/paper/2203.13591/) (2022)
- [Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervision](https://mlsanity.com/paper/1611.09813/) (2016)
- [Unsupervised Label Noise Modeling and Loss Correction](https://mlsanity.com/paper/1904.11238/) (2019)
- [Joint Optimization Framework for Learning with Noisy Labels](https://mlsanity.com/paper/1803.11364/) (2018)
- [CLIFF: Carrying Location Information in Full Frames into Human Pose and Shape Estimation](https://mlsanity.com/paper/2208.00571/) (2022)
- [Robust Test-Time Adaptation in Dynamic Scenarios](https://mlsanity.com/paper/2303.13899/) (2023)
- [Real-time self-adaptive deep stereo](https://mlsanity.com/paper/1810.05424/) (2018)
- [ViDA: Homeostatic Visual Domain Adapter for Continual Test Time Adaptation](https://mlsanity.com/paper/2306.04344/) (2023)
- [From Synthetic to Real: Unsupervised Domain Adaptation for Animal Pose Estimation](https://mlsanity.com/paper/2103.14843/) (2021)
- [Learning to Adapt for Stereo](https://mlsanity.com/paper/1904.02957/) (2019)
- [Bilevel Online Adaptation for Out-of-Domain Human Mesh Reconstruction](https://mlsanity.com/paper/2103.16449/) (2021)
- [Inference Stage Optimization for Cross-scenario 3D Human Pose Estimation](https://mlsanity.com/paper/2007.02054/) (2020)
- [Video Test-Time Adaptation for Action Recognition](https://mlsanity.com/paper/2211.15393/) (2022)
- [Meta-Tracker: Fast and Robust Online Adaptation for Visual Object Trackers](https://mlsanity.com/paper/1801.03049/) (2018)
- [Out-of-Domain Human Mesh Reconstruction via Dynamic Bilevel Online Adaptation](https://mlsanity.com/paper/2111.04017/) (2021)
- [On the Road to Online Adaptation for Semantic Image Segmentation](https://mlsanity.com/paper/2203.16195/) (2022)
- [Cyclic Test-Time Adaptation on Monocular Video for 3D Human Mesh Reconstruction](https://mlsanity.com/paper/2308.06554/) (2023)
- [Domain Adaptive 3D Pose Augmentation for In-the-wild Human Mesh Recovery](https://mlsanity.com/paper/2206.10457/) (2022)
- [On the Calibration of Human Pose Estimation](https://mlsanity.com/paper/2311.17105/) (2023)
- [KITRO: Refining Human Mesh by 2D Clues and Kinematic-tree Rotation](https://mlsanity.com/paper/2405.19833/) (2024)
- [Improving the Robustness of 3D Human Pose Estimation: A Benchmark and Learning from Noisy Input](https://mlsanity.com/paper/2312.06797/) (2023)

## Built on by (most cited first, 1 of 1)

- [Robust Long-term Test-Time Adaptation for 3D Human Pose Estimation through Motion Discretization](https://mlsanity.com/paper/2511.18851/) (2025)

## Abstract (arXiv)

Online test-time adaptation for 3D human pose estimation is used for video streams that differ from training data. Ground truth 2D poses are used for adaptation, but only estimated 2D poses are available in practice. This paper addresses adapting models to streaming videos with estimated 2D poses. Comparing adaptations reveals the challenge of limiting estimation errors while preserving accurate pose information. To this end, we propose adaptive aggregation, a two-stage optimization, and local augmentation for handling varying levels of estimated pose error. First, we perform adaptive aggregation across videos to initialize the model state with labeled representative samples. Within each video, we use a two-stage optimization to benefit from 2D fitting while minimizing the impact of erroneous updates. Second, we employ local augmentation, using adjacent confident samples to update the model before adapting to the current non-confident sample. Our method surpasses state-of-the-art by a large margin, advancing adaptation towards more practical settings of using estimated 2D poses.