发表机构
University of Electronic Science and Technology of China; King Fahd University of Petroleum & Minerals（电子科技大学; 法赫德国王石油与矿产大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出两阶段可变形卷积框架，结合监督初始化与对抗优化，从80维吸收光谱重构金属-绝缘体-金属谐振器几何，在多指标上优于多种卷积方法，实现了更优的光谱条件下几何重构。
AI 中文摘要
数据驱动的逆设计可高效生成具有指定光学响应的纳米光子结构，但由于非唯一性和精细几何特征，光谱到几何的映射仍具挑战性。本研究提出一种两阶段可变形卷积框架，用于从80维吸收光谱重构金属-绝缘体-金属谐振器几何结构。光谱被投影为150×4×4的潜在表示，并解码为64×64的谐振器掩码。训练结合了监督重构与最小二乘对抗优化，优化过程从最优监督检查点初始化。通过三次运行的消融实验，在相同架构下比较了可变形卷积与普通卷积、内卷积（involution）、动态卷积（Dynamic Conv）及ODConv的性能。所提模型取得了20.79±0.31 dB的峰值信噪比（PSNR）和0.8501±0.0082的结构相似性指数（SSIM），较普通卷积分别提升了2.16 dB和0.0831；还取得了0.9623±0.0027的戴斯系数（Dice）、0.9342±0.0038的交并比（IoU）和0.9550±0.0027的边界F值。采用冻结的前向替代模型评估光谱一致性，得到均方根误差（RMSE）为0.0805±0.0013，决定系数（R²）为0.7923±0.0065。学习到的偏移量在解码器的粗粒度和中间阶段表现出更强的自适应采样能力。总体而言，结合监督初始化与对抗优化的可变形采样方法，提升了光谱条件下的几何重构性能。
英文摘要
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.
