Leaving No OOD Instance Behind: Instance-Level OOD Fine-Tuning for Anomaly Segmentation

Yuxuan Zhang (Beijing University of Posts and Telecommunications) · Zhenbo Shi (University of Science and Technology of China) · han ye (University of Science and Technology of China) · Shuchang Wang (University of Science and Technology of China) · Zhidong Yu (University of Science and Technology of China) · Shaowei Wang (Guangzhou University) · Wei Yang (University of Science and Technology of China)

adaptive paradigmanomaly segmentationcomponent-level resultscomprehensive anomaly segmentationfeature separation objectiveglobal-level objectivesin-distribution surroundingsinference costinstance-level objectiveslnoib frameworkout-of-distribution fine-tuningrepresentation constraintssegmentation modelssmall anomaliestheoretical analysis

Out-of-distribution (OOD) fine-tuning has emerged as a promising approach for anomaly segmentation. Current OOD fine-tuning strategies typically employ global-level objectives, aiming to guide segmentation models to accurately predict a large number of anomaly pixels. However, these strategies often perform poorly on small anomalies. To address this issue, we propose an instance-level OOD fine-tuning framework, dubbed LNOIB (Leaving No OOD Instance Behind). We start by theoretically analyzing why global-level objectives fail to segment small anomalies. Building on this analysis, we introduce a simple yet effective instance-level objective. Moreover, we propose a feature separation objective to explicitly constrain the representations of anomalies, which are prone to be smoothed by their in-distribution (ID) surroundings. LNOIB integrates these objectives to enhance the segmentation of small anomalies and serves as a paradigm adaptable to existing OOD fine-tuning strategies, without introducing additional inference cost. Experimental results show that integrating LNOIB into various OOD fine-tuning strategies yields significant improvements, particularly in component-level results, highlighting its strength in comprehensive anomaly segmentation.