SegEarth-OV: Towards Training-Free Open-Vocabulary Segmentation for Remote Sensing Images

TL;DR

This paper introduces a training-free open-vocabulary segmentation method for remote sensing images, utilizing a novel upsampler and token bias correction, achieving significant improvements across multiple datasets.

Contribution

Proposes SimFeatUp, a training-free upsampler, and a token subtraction technique to enhance open-vocabulary segmentation in remote sensing images, addressing low-resolution and boundary issues.

Findings

Achieves up to 15.3% improvement over state-of-the-art methods.

Effective across diverse remote sensing tasks and datasets.

Introduces training-free techniques suitable for practical applications.

Abstract

Remote sensing image plays an irreplaceable role in fields such as agriculture, water resources, military, and disaster relief. Pixel-level interpretation is a critical aspect of remote sensing image applications; however, a prevalent limitation remains the need for extensive manual annotation. For this, we try to introduce open-vocabulary semantic segmentation (OVSS) into the remote sensing context. However, due to the sensitivity of remote sensing images to low-resolution features, distorted target shapes and ill-fitting boundaries are exhibited in the prediction mask. To tackle this issue, we propose a simple and general upsampler, SimFeatUp, to restore lost spatial information in deep features in a training-free style. Further, based on the observation of the abnormal response of local patch tokens to [CLS] token in CLIP, we propose to execute a straightforward subtraction operation to alleviate the global bias in patch tokens. Extensive experiments are conducted on 17 remote sensing datasets spanning semantic segmentation, building extraction, road detection, and flood detection tasks. Our method achieves an average of 5.8%, 8.2%, 4.0%, and 15.3% improvement over state-of-the-art methods on 4 tasks. All codes are released. \url{https://earth-insights.github.io/SegEarth-OV}