Neural Field Thermal Tomography:
A Differentiable Physics Framework for Non-Destructive Evaluation

Overview

A high-speed camera measures time-resolved transient surface temperature variations following localized heating with a pulsed laser. NeFTY utilizes a physics-aware framework to use these transient measurements to reconstruct the 3D subsurface diffusivity field and reveal hidden defects.

Interactive Reconstruction Results

Ground Truth

NeFTY (Ours)

U-Net (Full)*

PINN

Grid Opt.

3D Diffusivity Field (Defects only)

Predicted Surface Temp Evolution

N/A (Supervised method uses direct volumetric mapping)

L1 Error Evolution

Reference Frame

Reference Frame

* Note: The U-Net (Full) baseline relies on ground-truth diffusivity fields and does not simulate the forward thermal evolution.

Methodology: Analysis-by-Synthesis Loop

We propose Neural Field Thermal Tomography (NeFTY), a differentiable physics framework for the quantitative 3D reconstruction of material properties from transient surface temperature measurements. NeFTY parameterizes the 3D diffusivity field as a continuous neural field optimized through a rigorous numerical solver. By leveraging a differentiable physics solver, our approach enforces thermodynamic laws as hard constraints while maintaining the memory efficiency required for high-resolution 3D tomography.

Citation

@article{zhong2026nefty,
  title={Neural Field Thermal Tomography: A Differentiable Physics Framework for Non-Destructive Evaluation},
  author={Zhong, Tao and Hu, Yixun and Zheng, Dongzhe and Sood, Aditya and Allen-Blanchette, Christine},
  journal={arXiv preprint arXiv:2603.11045},
  year={2026}
}

Contact: Tao Zhong (tzhong@princeton.edu) | CAB Lab, Princeton University