Meta-Learned Adaptive Optimization for Robust Human Mesh Recovery with Uncertainty-Aware Parameter Updates
Shaurjya Mandal, Nutan Sharma, John Galeotti · 2026
Read Meta-Learned Adaptive Optimization for Robust Human Mesh Recovery with Uncertainty-Aware Parameter Updates by Shaurjya Mandal, Nutan Sharma, John Galeotti (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-26447.txt.
Description
Mandal, Sharma, and Galeotti propose a meta-learning method for rebuilding a human body mesh from one photo. Depth ambiguity and domain shifts make the task hard, and test-time refinement often starts from a poor fit. Their training mimics test-time tuning so the model learns starts that refine well. A caching step freezes joints that have settled, cutting wasted work. Updates drawn from learned distributions add robustness and a measure of doubt. The method cuts MPJPE by 10.3 points on 3DPW.
Catalog record
| CREATOR | Shaurjya Mandal, Nutan Sharma, John Galeotti |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 39878 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.26447