Clad personal blog
A 3D Body from Eight Questions – No Photo, No GPU
We trained an MLP that predicts 58 body shape parameters from 8 questionnaire inputs. The trick: a differentiable physics loss. Mass accuracy: 0.3 kg mean vs 3.9 kg for Ridge. No GPU at inference.
- Points
- 139
- HN comments
- 43
- Submissions
- 1
- Cited by
- 0 commenters
AI & Machine LearningDesign & UX2026
On Hacker News
| Submitted | Title on HN | Points | Comments |
|---|---|---|---|
| Apr 22, 2026 | A 3D Body from Eight Questions – No Photo, No GPU by arkadiuss | 139 | 43 |
