HADD: A hierarchy-aware disentangled diffusion framework with spatio-temporal denoising for 3D human pose estimation
Jinze Jiang, Zhongheng Jian, Hualing Zheng, Minsi Liang, Yuxing Shi, Liming Zhang, Gennady S Cymbalyuk · 2026
Read HADD: A hierarchy-aware disentangled diffusion framework with spatio-temporal denoising for 3D human pose estimation by Jinze Jiang, Zhongheng Jian, Hualing Zheng, Minsi Liang, Yuxing Shi, Liming Zhang, Gennady S Cymbalyuk (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/PMC13619107.txt.
Description
Jinze Jiang and six colleagues introduce HADD, a method for estimating three-dimensional human poses from a single camera view. Their framework splits a pose into bone lengths and bone directions, adds noise to each part, and then removes that noise step by step. A denoising module built around the skeleton's hierarchy is meant to stop errors from growing at joints far from the body's center.
Experiments used two large motion datasets, Human3.6M and MPI-INF-3DHP, which record people moving in laboratory settings. On the second dataset the method reached a mean joint error of under thirty millimeters with very high accuracy scores under a single-guess setting. The authors present the gains as help for sports training, rehabilitation checks, and animation work that need fine motion detail.
Catalog record
| CREATOR | Jinze Jiang, Zhongheng Jian, Hualing Zheng, Minsi Liang, Yuxing Shi, Liming Zhang, Gennady S Cymbalyuk |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 128 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |