A single consumer-grade GPU running a vision-language model (VLM) is used to supply missing guidance on pixel-level segmentation, improving segmentation while producing structured, auditable evidence that drives the result and can be inspected on its own.
Abstract
Global welfare often depends on the correct interpretation of aerial and satellite imagery. Acting on such imagery (mapping flooded ground, crop extent, or damaged infrastructure) demands pixel-level segmentation to ensure perfect class localization. Pretrained general foundation models, when applied directly, often miss important features and cannot always find all the classes belonging to a given scene, overlooking smaller objects that matter most. We use a single consumer-grade GPU running a vision-language model (VLM) to supply this missing guidance, improving segmentation while producing structured, auditable evidence that drives the result and can be inspected on its own. We fuse three approaches: the frozen foundation model that labels every pixel, and two queries to a VLM, one to choose the classes that matter, and one to locate the small objects the base model misses. Evaluating across four aerial datasets, we see consistent gains at each stage where the base model is competent.
DLPANet is proposed, a novel dual-level prototype alignment network centered on Prototype-Guided Spatial Attention, enabling simultaneous modeling of scene context and fine-grained details and demonstrates that the decoupled dual cross-attention mechanism provides superior prototype-query alignment compared to prior gl...
Mustafa Alawadi, M. Fateh· Jordanian Journal of Compute...· 0 citations
The results show that VLM image tokens provide useful semantic cues for transferring driving attributes to objects outside the training vocabulary, and this model matches strong vision-only segmenters on familiar categories and improves transfer to real open-world anomalies.
Yu-Chen Zhang, Yuan Gao, Sebastian Schmidt et al.· 0 citations
Autonomous vehicles operating in open-world scenarios are inevitably confronted with previously unknown objects, such as exotic animals or loose cargo. The reliable detection and segmentation of these out-of-distribution (OOD) objects is therefore crucial for a safe understanding of the environment and decision-making....
Serin Varghese, Fabian Hüger, Kira Maag· 0 citations
Due to extreme density variations and scarce visual features, tiny-object detection in low-altitude uncrewed aerial vehicle (UAV) images remains a challenging task. Although DETR-based detectors have shown promising potential, they rely on a fixed number of object queries, which leads to suboptimal performance in dense...
Ling-Feng Lin, Bo-Xiang Xie, W. Luo et al.· IEEE Geoscience and Remote S...· 0 citations
Large vision models provide useful representations for remote-sensing segmentation but are often too expensive for deployment at the satellite or field edge. Existing feature-level distillation methods also tend to assume similar teacher and student architectures and often stop feature alignment when task training begi...
Kishor Kumar Bhaumik, Nícolas Roque dos Santos, Jia Chen et al.· 0 citations
Marigold V2 achieves state-of-the-art results when applied to other dense regression tasks, such as surface normals estimation and intrinsic image decomposition, and achieves state-of-the-art results when applied to other dense regression tasks, such as surface normals estimation and intrinsic image decomposition.
Igor Pavlovic, Thiemo Wandel, Anton Obukhov et al.· 1 citation· ⚡1
