Search papers, labs, and topics across Lattice.
This paper introduces OVEarth-Bench, a novel benchmark for open-vocabulary Earth observation that enhances evaluation through broader category coverage and diverse query types. The evaluation of various methods reveals that while current approaches show limited performance, MLLM-based models outperform others, and EO-specific methods generally lag behind general models. These insights underscore the necessity for more comprehensive and realistic benchmarks to improve the design of future open-vocabulary EO methods.
MLLM-based methods outperform traditional EO-specific approaches, revealing a critical gap in current Earth observation evaluations.
Open-vocabulary Earth observation (EO) aims to localize geospatial concepts specified in natural language rather than a fixed label set. Existing benchmarks, however, usually cover narrow category vocabularies or limited query forms. To fill this gap, we introduce OVEarth-Bench, which extends existing evaluation in two directions: category breadth, through broad hierarchical category coverage with positive and negative expressions, and query diversity, through vocabulary, referring, and reasoning queries. The benchmark supports mask and box localization under a unified zero-shot protocol. We evaluate a broad set of general and EO-specific methods. The evaluation reveals that: (1) the performance of current methods remains limited, while broader category coverage yields more stable model rankings; (2) MLLM-based methods achieve the strongest overall performance; and (3) EO-specific methods generally underperform general models and rarely match the strongest methods. These findings provide guidance for future open-vocabulary EO method design and highlight the importance of developing more realistic, diverse, high-quality, and large-scale benchmarks for reliable evaluation. Our data and evaluation package are released at https://earth-insights.github.io/OVEarth-bench.
