Context-measure: Contextualizing Metric for Camouflage
This provides a foundational evaluation benchmark for computer vision applications involving camouflaged patterns in domains like agriculture, industry, and medicine.
The paper tackles the problem that existing camouflage evaluation metrics ignore spatial context dependencies, proposing Context-measure as a new contextualized evaluation paradigm based on probabilistic pixel-aware correlation. Experiments on three camouflaged object segmentation datasets show it delivers more reliability than context-independent metrics.
Camouflage is primarily context-dependent yet current metrics for camouflaged scenarios overlook this critical factor. Instead, these metrics are originally designed for evaluating general or salient objects, with an inherent assumption of uncorrelated spatial context. In this paper, we propose a new contextualized evaluation paradigm, Context-measure, built upon a probabilistic pixel-aware correlation framework. By incorporating spatial dependencies and pixel-wise camouflage quantification, our measure better aligns with human perception. Extensive experiments across three challenging camouflaged object segmentation datasets show that Context-measure delivers more reliability than existing context-independent metrics. Our measure can provide a foundational evaluation benchmark for various computer vision applications involving camouflaged patterns, such as agricultural, industrial, and medical scenarios. Code is available at https://github.com/pursuitxi/Context-measure.