Semantic geometric decoupling improves logical anomaly detection accuracy
SGDB: Semantic-Geometric Decoupling and Evidence Binding for Logical Anomaly Detection
Computer Vision and Pattern Recognition
Summary
Logical anomaly detection means finding when something in a scene does not follow expected rules, like parts being the wrong size or shape. The paper’s authors created a method called SGDB that separates understanding what objects are from measuring their shapes and sizes to avoid mixing these up. Their method carefully links shape evidence to the right objects and uses rules based on normal examples to decide if something is wrong. This approach also explains its decisions clearly and works well on complex benchmarks.
What this means in practice
- •For industrial quality control teams: Detect logical configuration errors in manufacturing parts to prevent faulty product assembly.
- •For robotics system developers: Improve robot scene understanding by verifying object identities and geometric relations match task expectations.
Authors
Chenrui Ji, Jiafan Lu, Zilu Luo, Feng Wan, Qiuhui Chen, Weimin Zhong
Abstract
Logical anomaly detection requires recognizing scene components and measuring whether their configuration satisfies task constraints. Correct identities can coexist with inaccurate geometry; clean mask fragments can yield misleading lengths. We propose Semantic-Geometric Decoupling and Evidence Binding (SGDB). A semantic decoder learns object identities from component-annotated normal sources and factual edits; a separate reader supplies final measurements. Partial one-to-one association binds geometric witnesses to identities while retaining unmatched objects in the count. A support-certified reader selects mask-derived axes using line quality and complete component ownership. Explicit task requirements and normal-data references define executable checks, with decision thresholds fixed by independent normal calibration. Execution-grounded explanation traces rule contributions to source objects and measurements, extracting compact sufficient alarm certificates. A frozen language model expresses selected outcomes as cited claims, with numeric evidence retained by the executor. Reports preserve observed deviations and unavailable evidence. SGDB achieves 98.19% AUROC and 94.52% fixed-threshold F1 on five MVTec LOCO logical categories, and 93.56% AUROC and 92.76% F1 across all 50 VID-AD tasks. Frozen-output controls evaluate geometric readout and explanatory sufficiency and compactness.