← Back to all news
arXiv Computer Vision

Vision-Language Model for Accurate Crater Detection

The paper presents a deep‑learning crater detection algorithm (CDA) based on the OWLv2 Vision Transformer, fine‑tuned with Low‑Rank Adaptation on a manually labeled IMPACT dataset. It optimizes a combined loss of CIoU for localization and contrastive loss for classification, achieving a maximum recall of 92.6% and precision of 71.4% on lunar images. The method demonstrates reliable crater detection under varied illumination and rugged terrain, supporting safer lunar landings.

Related stories

arXiv Computer Vision
Sep 3

Adapting a Foundation Model for Lunar Surface Height Estimation

The paper proposes a method to adapt the Depth Anything V2 (DAV2) zero‑shot relative depth model for estimating lunar surface height. By fine‑tuning DAV2 with publicly available stereophotogrammetry‑derived DEM data, the authors achieve a significant performance boost over the unadapted zero‑shot model. This improved estimator can provide more accurate relative height information useful for hazard detection in future ESA lunar landings.

By Patrick Bauer, Marius Schwinning, Melanie Siegel, Andreas Weinmann, Hichem Snoussi
computer-visionfine-tuningbenchmarks
More like this →
arXiv AI
Sep 15

Multimodal-Multiresolution Foundation Model for Lunar Remote Sensing

The paper introduces a multimodal foundation model for lunar remote sensing, trained from scratch on SomBench—a dataset of nearly two million co‑registered tile bundles across 11 modalities at 1 m and 100 m resolutions. The model extends the TerraMind masked‑token architecture with lunar‑specific features such as explicit acquisition geometry and joint training of two spatial scales, and employs FlexiViT patch embeddings for adaptable patch sizes. Evaluation on crater detection, irregular mare patch segmentation, and polar ice prospectivity regression shows that the pretrained model matches or surpasses ImageNet‑pretrained baselines, with notable label efficiency and effective adaptation via LoRA.

By Paolo Fraccaro, Gabby Nyirjesy, Daniela Szwarcman, Himanshu Patil, Vishal Gaur, Rohit Lal, Rachel A. Slank, Geoffrey Dawson, Hiyam Debary, Michael K. Barker, Andrew Annex, Vishnu Viswanathan, Zachary Morse, Ethan I. Schaefer, Nikolaos Dionelis, Ankur Kumar, Campbell D. Watson, Manil Maskey, Rebekah I. Dawson-Rigas, Juan Bernab\'e-Moreno, Rahul Ramachandran, Sujit Roy
ragcomputer-visionfine-tuningmultimodalbenchmarks
More like this →
arXiv AI
Jul 16

Post-Disaster Affected Area Segmentation with a Vision Transformer (ViT)-based EVAP Model using Sentinel-2 and Formosat-5 Imagery

arXiv:2507. 16849v3 Announce Type: replace-cross Abstract: We propose a vision transformer (ViT)-based deep learning framework to refine disaster-affected area segmentation from remote sensing imagery, aiming to support and enhance the Emergent Value Added Product (EVAP) developed by the Taiwan Space Agency (TASA).

By Yi-Shan Chu, Hsuan-Cheng Wei
llmscomputer-vision
More like this →
arXiv Computer Vision
Sep 15

Global-Local Contextual Progressive Expansion Network for Martian Landslide Segmentation in Multimodal Remote Sensing Imagery

arXiv:2609.13332v1 Announce Type: new Abstract: Automated landslide segmentation on Mars is one of the important tasks for understanding its surface processes, and all will aid in future space explor...

By Leo Thomas Ramos, Sidike Paheding, Abel A. Reyes-Angulo, Rajaneesh A., Sajinkumar K. S., Angel D. Sappa, Thomas Oommen
llmscomputer-visionmultimodalbenchmarks
More like this →
arXiv Computer Vision
Oct 7

MoonGS: High-quality Representation of the Lunar Surface via Gaussian Splatting Using Robust Depth Features from Image Pairs

arXiv:2610.07110v1 Announce Type: new Abstract: High-quality 3D reconstruction of lunar terrain from sparse rover images is indispensable for autonomous lunar exploration, but remains challenging bec...

By Yun Jiang, Bo Zheng, Yingying Zhang, Xueming Xiao, Tao Hu, Hutao Cui, Zhiguo Meng, Ke Gao, Yang Gao, Meibao Yao
agentsbenchmarks
More like this →
arXiv AI
Sep 17

Tabular Deep Learning vs Classical Machine Learning for Urban Land Cover Classification

The paper compares classical machine learning algorithms—such as Logistic Regression, SVM, Random Forest, XGBoost, and CatBoost—with Tabular Deep Learning models (TabNet, FT-Transformer, TabTransformer, TabSeq, and 1D CNNs) for urban land cover classification using a UCI dataset derived from high‑resolution aerial imagery. It evaluates performance across nine land cover classes, addressing challenges like high dimensionality, heterogeneous features, and class imbalance by applying weighted cross‑entropy loss for deep models and measuring accuracy, macro‑precision, macro‑recall, macro‑F1, AUC‑ROC, and confusion matrices. Results indicate that while tree ensembles remain strong baselines, Tabular Deep Learning can match or surpass them when non‑linear interactions are prominent and imbalance handling is effective.

By Muntasir Tabasum, Tanpia Tasnim, Md. Ekramul Islam, Al Zadid Sultan Bin Habib
llmsbenchmarks
More like this →