Search papers, labs, and topics across Lattice.
This study introduces a two-stage deformable-convolutional framework for the inverse design of nanophotonic absorbers, effectively mapping 80-dimensional absorption spectra to resonator geometries. By employing a combination of supervised reconstruction and least-squares adversarial refinement, the model achieves significant improvements in image quality metrics such as PSNR and SSIM compared to traditional convolution methods. The results demonstrate that adaptive sampling enhances the reconstruction of complex geometries, leading to a high degree of spectral consistency and accuracy in the generated designs.
Deformable convolution outperforms traditional methods, achieving a remarkable 20.79 dB PSNR and 0.9623 Dice score in nanophotonic absorber design.
Data-driven inverse design enables efficient generation of nanophotonic structures with prescribed optical responses, but spectrum-to-geometry mapping remains challenging due to non-uniqueness and fine geometric features. This work presents a two-stage deformable-convolutional framework for reconstructing metal--insulator--metal resonator geometries from 80-dimensional absorption spectra. The spectrum is projected to a $150\times4\times4$ latent representation and decoded into a $64\times64$ resonator mask. Training combines supervised reconstruction with least-squares adversarial refinement initialized from the best supervised checkpoint. A three-run ablation compares deformable convolution with plain convolution, involution, Dynamic Conv, and ODConv under the same architecture. The proposed model achieves $20.79\pm0.31$~dB PSNR and $0.8501\pm0.0082$ SSIM, improving over plain convolution by 2.16~dB and 0.0831, respectively. It further achieves Dice $0.9623\pm0.0027$, IoU $0.9342\pm0.0038$, and boundary F-score $0.9550\pm0.0027$. Spectral consistency evaluated using a frozen forward surrogate yields RMSE $0.0805\pm0.0013$ and $R^2=0.7923\pm0.0065$. Learned offsets show stronger adaptive sampling at coarse and intermediate decoder stages. Overall, deformable sampling with supervised initialization and adversarial refinement improves spectrum-conditioned geometry reconstruction.
