arXiv:2609.25500v1 Announce Type: new Abstract: Training data quantity and quality greatly affect object detection model performance, regardless of model architecture. When using object detection mod...
The paper presents an AI-based system for onboard maritime object detection on Earth‑observation satellites, aiming to reduce data volume by selectively downlinking relevant vessel information. A YOLOX‑S detector is trained on a dataset of 68 Maxar scenes with 43 vessel classes, then quantized and deployed on a Versal VC1902 DPU with minimal loss in accuracy and a few‑second processing time per scene. Various downlink modes—metadata only, image crops, tiles, or full scenes—are evaluated, showing that tiles retain 98% of vessels while reducing scene volume to 29%, and crops retain 83% with only 3% of the volume.
Vessel perception from space is crucial for a wide range of maritime applications, from traffic monitoring to environmental protection. However, most existing datasets predominantly focus on general o...
arXiv:2609.37003v1 Announce Type: new Abstract: Vessel perception from space is crucial for a wide range of maritime applications, from traffic monitoring to environmental protection. However, most e...
CoralscapesV2 is an expanded dataset for coral reef visual scene understanding, increasing the number of fine‑grained classes from 39 to 95 and adding 65,000 exhaustive fish instance masks. It supports panoptic segmentation by providing high‑quality semantic and instance labels across diverse, unconstrained reef imagery. The dataset serves as a challenging benchmark for modern segmentation models and enables broader applications such as benthic cover mapping and automated fish‑reef interaction analysis.
arXiv:2608. 15790v1 Announce Type: new Abstract: Crevasse mapping from uncrewed aerial vehicle (UAV) imagery matters for glaciological research and for field safety in glaciated terrain.
The paper introduces AquaOV255, a large-scale underwater segmentation dataset with 255 categories and over 20,000 images, and establishes the UOVSBench benchmark by combining it with five other datasets. It also presents Earth2Ocean, a training‑free open‑vocabulary segmentation framework that adapts terrestrial vision‑language models to underwater scenes using a Geometric‑guided Visual Mask Generator and a Category‑visual Semantic Alignment module. Experiments on UOVSBench show that Earth2Ocean significantly improves performance while keeping inference efficient.
The paper introduces a pipeline that localizes frames from historical PTZ maritime video onto a reference panorama and uses context-aware sampling to build compact, scene‑specific training sets for ship detection. By enriching frames with weather and solar metadata and applying diversity sampling, the method reduces 40,718 candidate frames to just 220 for annotation, achieving a 99.5% reduction. Fine‑tuned YOLOv26‑m on this curated subset attains high detection performance (AP50 ≈ 94.8% and AP50‑95 ≈ 75.1%).
The paper introduces a pipeline that localizes frames from historical PTZ maritime video onto a reference panorama and then selects a context‑aware, diverse subset for ship detection training. By combining SuperPoint‑LightGlue localization, weather and solar‑state metadata, and diversity sampling, the method reduces 40,718 candidate frames to just 220 images for annotation. Fine‑tuned YOLOv26 on this compact set achieves high detection performance (AP50 ≈ 94.8%) while cutting annotation effort by 99.5%.
The paper presents a weakly supervised semantic segmentation approach for mapping seagrass habitats using side‑scan sonar imagery. By training a ViT‑based encoder‑decoder with image‑level labels, class activation maps are refined into pseudo‑labels and iteratively self‑trained, achieving high mean intersection‑over‑union scores (up to 89.3 %) without pixel‑level annotations. The method also benefits from self‑supervised pretraining and demonstrates generalizability in field trials.
The paper presents a method that leverages sparse expert point annotations from historical benthic surveys to improve dense segmentation of marine imagery. By using these points as visual prompts for the SAM2 foundation model and introducing a mechanism to filter out unreliable points, the authors generate high‑quality pseudo‑ground‑truth masks that train more accurate fine‑grained semantic segmentation models. The approach is validated on public benthic datasets and a new benchmark featuring real‑world sparse annotations, aiming to enable scalable ecological analysis.
The paper introduces PhySeg, a SAM-based segmentation framework that uses a lightweight refinement module to progressively restore fine-grained details via multi-scale feature fusion, improving phytoplankton segmentation accuracy. It also presents a newly annotated phytoplankton dataset with instance-level masks, offering a valuable resource for future research. Experimental results confirm the effectiveness of the proposed module, and the code is publicly available on GitHub.