发表机构
University of South Florida; University of Central Florida（南佛罗里达大学; 中佛罗里达大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出DARB和ARS，通过非对称物理先验与自适应监督，实现一个双向声呐-光学转换模型，性能接近或超越专用模型。
AI 中文摘要
在成像声呐与光学相机之间进行转换对于水下感知具有重要意义，但使用独立模型支持两个方向会导致存储和计算资源的重复。因此，一个统一的双向模型颇具吸引力，然而现有方法大多将两个方向对称处理，尽管它们在图像形成物理机制上存在根本差异。我们认为，共享生成模型并不需要共享物理机制。我们提出了方向非对称现实桥（DARB），该模型保留共享的扩散桥主干，同时通过非对称路径路由方向特定的物理先验：用于声呐到光学转换的距离感知调制，以及用于光学到声呐转换的极坐标射线相关处理。我们进一步表明，训练中的对称性同样代价高昂：应用统一的现实调度会使声呐到光学的峰值信噪比（PSNR）降低2.60 dB。我们的自适应现实监督（ARS）则根据重建质量和梯度平衡来决定何时、何地以及以何种强度应用感知监督。DARB和ARS共同使一个双向模型在PSNR上达到与声呐到光学专用模型相差0.11 dB以内的性能，在FID上比光学到声呐专用模型提升0.70，并在八项指标中的七项上超越两个独立训练的BBDM。
英文摘要
Translating between imaging sonar and optical cameras is valuable for underwater perception, but supporting both directions with separate models duplicates storage and computation. A unified bidirectional model is therefore attractive, yet existing approaches largely treat the two directions symmetrically despite their fundamentally different image-formation physics. We argue that sharing a generative model does not require sharing the physics. We introduce the Direction-Asymmetric Realism Bridge (DARB), which retains a shared diffusion-bridge trunk while routing direction-specific physical priors through asymmetric pathways: range-aware modulation for sonar-to-optical translation and polar ray-dependent processing for optical-to-sonar translation. We further show that symmetry in training is also costly: applying a common realism schedule reduces sonar-to-optical PSNR by 2.60 dB. Our Adaptive Realism Supervision (ARS) instead determines when, where, and how strongly perceptual supervision is applied from reconstruction quality and gradient balance. Together, DARB and ARS enable one bidirectional model to match the sonar-to-optical specialist within 0.11 dB PSNR, outperform the optical-to-sonar specialist by 0.70 FID, and surpass two independently trained BBDMs on seven of eight metrics.
Comments10 pages, 5 figures, 3 tables
