SonarLLM is a multimodal large language model that treats sonar as a native perceptual modality, combining a sonar‑specific encoder, physics‑aware feature enhancement, and reliability‑aware hierarchical fusion to align acoustic structure with optical semantics. The authors introduce SonarBench, a benchmark covering recognition, counting, visual question answering, and captioning across sonar‑only, optical‑only, and fusion settings, enabling controlled measurement of cross‑modal complementarity. SonarLLM achieves 72.0% macro accuracy on sonar‑only tasks and 68.7% under fusion, outperforming baselines by significant margins and demonstrating increasing fusion gains as optical visibility degrades.
By Cong Su, longxuan ma, Ling Dong, Guofeng Tang, Weijie Yin, Haohui Chen, Zhengtao Yu
The paper introduces SASC-USOD, a framework for underwater salient object detection that learns spatially adaptive coordination between two structural representations: a boundary-sensitive representation using Laplacian filtering and a region-coherent representation via dual-range anisotropic large-kernel aggregation. A spatial coordination module estimates the relative reliability of these representations and adaptively blends them based on image content. Experiments on USOD10K and USOD benchmarks show that SASC-USOD outperforms existing methods, reducing MAE by 4.07% and 23.53% respectively, and its lightweight variant achieves 21 FPS on an NVIDIA Jetson TX2 NX.
By Lin Hong, Chenhui Wang, Linan Deng, Yuning Cui, Yu Zhang, Xin Wang, Bojian Zhang, Xingchen Yang, Fumin Zhang
arXiv:2608.23215v1 Announce Type: cross
Abstract: Automated perception in side-scan sonar (SSS) imagery is severely hindered by physical acoustic artifacts, resulting in representations that inextric...
By Taqi Hamoda, Hayat Rajani, Nuno Gracias
Infrared-visible image fusion (IVIF) is pivotal for multimodal perception, yet reconciling the inherent information disparity between thermal and textural features remains a fundamental challenge. Existing prior-guided methods often rely on static constraints that induce optimization conflicts or utilize extrinsic semantic priors from large-scale foundation models (e.
arXiv:2608.25367v1 Announce Type: new
Abstract: Underwater unimodal object detection faces many challenges in sensor imaging, such as optical images limited by underwater noise and visible distance,...
By Zhuoyan Liu, Yihan Wang, Bo Wang, Bing Wang, Ye Li
arXiv:2604.00313v3 Announce Type: replace
Abstract: Underwater image classification is constrained by the cost of annotation and by the computational and methodological requirements of task-specific...
By Thomas Manuel Rost, Martina Figlia, F. Morgado-Dias, Marko Radeta
arXiv:2608.22888v1 Announce Type: new
Abstract: Reconstructing photorealistic scenes in unconstrained underwater environments remains challenging due to severe media-induced light scattering and unpr...
By Xiaopeng Guo, Wai Chung Tse, Yipeng Zhu, Hanwen Zhang, Huajian Huang, Sai-Kit Yeung
arXiv:2608.23479v1 Announce Type: new
Abstract: Side-Scan Sonar (SSS) is a primary modality for large-scale underwater mapping, yet automated perception and cross-modal alignment are severely bottlen...
By Taqi Hamoda, Nuno Gracias
arXiv:2602. 24181v2 Announce Type: replace-cross Abstract: Pre-trained vision encoders like DINOv2 have demonstrated exceptional performance on unimodal tasks.
By Rishabh Kabra, Maks Ovsjanikov, Drew A. Hudson, Ye Xia, Skanda Koppula, Andre Araujo, Joao Carreira, Niloy J. Mitra
arXiv:2607. 06882v1 Announce Type: cross Abstract: Visual navigation policies built on large pretrained models have so far followed a common recipe: a dedicated visual encoder, a bespoke action head, and training on thousands of hours of cross-embodiment datasets.
By Peter Bohm, Saimunur Rahman, Abdelwahed Khamis, Sagun Man Singh Shrestha, Chris McCool, Peyman Moghadam
arXiv:2510. 04876v3 Announce Type: replace-cross Abstract: Benthic habitat mapping is fundamental for understanding marine ecosystems, guiding conservation efforts, and supporting sustainable resource management.
By Hayat Rajani, Valerio Franchi, Borja Martinez-Clavel Valles, Raimon Ramos, Rafael Garcia, Nuno Gracias
Visual navigation policies built on large pretrained models have so far followed a common recipe: a dedicated visual encoder, a bespoke action head, and training on thousands of hours of cross-embodiment datasets. We ask whether this recipe is necessary.