Component-Based Out-of-Distribution Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Wenrui, Chang, Hong, Hou, Ruibing, Shan, Shiguang, Chen, Xilin
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910160414310400
author Liu, Wenrui
Chang, Hong
Hou, Ruibing
Shan, Shiguang
Chen, Xilin
author_facet Liu, Wenrui
Chang, Hong
Hou, Ruibing
Shan, Shiguang
Chen, Xilin
contents Out-of-Distribution (OOD) detection requires sensitivity to subtle shifts without overreacting to natural In-Distribution (ID) diversity. However, from the viewpoint of detection granularity, global representation inevitably suppress local OOD cues, while patch-based methods are unstable due to entangled spurious-correlation and noise. And neither them is effective in detecting compositional OODs composed of valid ID components. Inspired by recognition-by-components theory, we present a training-free Component-Based OOD Detection (CoOD) framework that addresses the existing limitations by decomposing inputs into functional components. To instantiate CoOD, we derive Component Shift Score (CSS) to detect local appearance shifts, and Compositional Consistency Score (CCS) to identify cross-component compositional inconsistencies. Empirically, CoOD achieves consistent improvements on both coarse- and fine-grained OOD detection.
format Preprint
id arxiv_https___arxiv_org_abs_2604_21546
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Component-Based Out-of-Distribution Detection
Liu, Wenrui
Chang, Hong
Hou, Ruibing
Shan, Shiguang
Chen, Xilin
Computer Vision and Pattern Recognition
Out-of-Distribution (OOD) detection requires sensitivity to subtle shifts without overreacting to natural In-Distribution (ID) diversity. However, from the viewpoint of detection granularity, global representation inevitably suppress local OOD cues, while patch-based methods are unstable due to entangled spurious-correlation and noise. And neither them is effective in detecting compositional OODs composed of valid ID components. Inspired by recognition-by-components theory, we present a training-free Component-Based OOD Detection (CoOD) framework that addresses the existing limitations by decomposing inputs into functional components. To instantiate CoOD, we derive Component Shift Score (CSS) to detect local appearance shifts, and Compositional Consistency Score (CCS) to identify cross-component compositional inconsistencies. Empirically, CoOD achieves consistent improvements on both coarse- and fine-grained OOD detection.
title Component-Based Out-of-Distribution Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.21546