Neural Fields for NV-Center Inverse Sensing

2026-05-13

AI Summary

This work addresses the challenge in nitrogen-vacancy (NV) center magnetometry where the nonlinear, spectrally coupled, and physically sensitive forward model renders conventional inversion methods ineffective. To overcome this, the authors propose NeTMY, a novel approach that for the first time integrates an amortization-free coordinate-based neural field with a differentiable NV physics model in an end-to-end framework. By incorporating annealed positional encoding, multi-scale optimization, a sparsity-aware gating mechanism, and a spectral fidelity loss, NeTMY effectively mitigates the central collapse problem inherent in free spin density optimization, enabling high-fidelity reconstruction of sparse spin sources. Experiments on synthetic data demonstrate that NeTMY outperforms existing methods in both localization accuracy and distribution metrics, confirming its efficacy in solving physically faithful neural inverse problems.

Abstract

Inverse problems in scientific sensing are often solved with either hand-designed regularizers or supervised networks trained on simulated labels, yet both can fail when the forward model is nonlinear, spectrally coupled, and physically delicate. We study this issue for noise sensing based on nitrogen-vacancy (NV) centers in diamond, where a quantum sensor measures magnetic-noise spectra generated by sparse spin sources. We show that replacing a common scalar/coherent forward approximation with a tensor power-summed dipolar operator changes the inverse landscape and exposes a center-collapse failure mode in free-density optimization. We propose NeTMY, an amortization-free coordinate neural field coupled to the differentiable NV forward model, with annealed positional encoding, multiscale optimization, sparsity/gating, and spectrum-fidelity losses. Across sparse synthetic reconstructions generated by the corrected operator, NeTMY achieves the best localization and distributional metrics in the tested benchmark. Mechanism experiments show that NeTMY does not directly execute the raw density-space gradient; its parameterization smooths and redistributes updates, mitigating the center-collapse pathology. These results position NV quantum sensing as a useful testbed for physics-faithful neural inverse problems.

Problem

• inverse sensing
• NV centers
• neural fields
• center-collapse
• magnetic-noise spectra

Innovation

• neural fields
• inverse sensing
• NV centers
• differentiable physics
• tensor dipolar operator