RS-OVC: Open-Vocabulary Counting for Remote-Sensing Data

2026-04-09

AI Summary

This work addresses the limitations of existing remote sensing image object counting methods, which are confined to predefined categories and struggle to generalize to novel classes without costly re-annotation and retraining. To overcome this, we propose RS-OVC, the first open-vocabulary counting model tailored for remote sensing imagery, thereby introducing open-vocabulary counting to this domain for the first time. RS-OVC leverages textual and/or visual conditioning to enable accurate zero-shot counting of categories unseen during training, eliminating the need for additional annotations. Experimental results demonstrate that RS-OVC achieves high counting accuracy on unseen classes, significantly enhancing the practicality and adaptability of remote sensing monitoring in dynamic real-world scenarios.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

Object-Counting for remote-sensing (RS) imagery is attracting increasing research interest due to its crucial role in a wide and diverse set of applications. While several promising methods for RS object-counting have been proposed, existing methods focus on a closed, pre-defined set of object classes. This limitation necessitates costly re-annotation and model re-training to adapt current approaches for counting of novel objects that have not been seen during training, and severely inhibits their application in dynamic, real-world monitoring scenarios. To address this gap, in this work we propose RS-OVC - the first Open Vocabulary Counting (OVC) model for Remote-Sensing and aerial imagery. We show that our model is capable of accurate counting of novel object classes, that were unseen during training, based solely on textual and/or visual conditioning.

Problem

• Open-Vocabulary Counting
• Remote-Sensing
• Object-Counting
• Novel Class Generalization

Innovation

• Open-Vocabulary Counting
• Remote-Sensing Imagery
• Zero-Shot Object Counting
• Text-to-Image Conditioning
• Aerial Image Analysis