AsymLoc: Asymmetric Point Supervision for Cross-Dataset Zero-Shot 3D Anomaly Localization
Sensors · 10.11.2026 · 26 (22) · 7109
DOI: 10.3390/s26227109 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 10.11.2026. Verfasst von Qianhui Wang und 8 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Pioneer and Leading Goose R&D Program of Zhejiang Province.
Abstract (Originalsprache)
Cross-dataset zero-shot 3D anomaly localization aims to identify defective regions in previously unseen object categories without accessing target-domain data during training. Although multi-view vision–language approaches provide transferable semantic representations, their point-level anomaly responses may remain insufficiently discriminative, and improving point-level localization does not necessarily improve instance-level discrimination. We propose AsymLoc, an asymmetric point supervision framework built upon multi-view projection with prompted Contrastive Language–Image Pre-training (MVP-PCLIP). The framework emphasizes anomalous points through positive supervision and uses selective hard-normal regularization as an auxiliary constraint on high-scoring normal regions. A decoupled instance-level score fusion module combines view-level semantic evidence with point-level anomaly responses while preventing sample-level gradients from rewriting the localization map. Under the MVTec 3D-AD→Real3D-AD protocol, AsymLoc improves point-level AUROC, F1 score, and average precision by 2.8, 7.6, and 7.0 points over a reproduced MVP-PCLIP baseline, and instance-level AUROC by 8.4 points. On the reverse Real3D-AD→MVTec 3D-AD direction, point-level metrics remain close to the matched baseline while instance-level AUROC rises by 3.7 points. Positive supervision accounts for most of the localization improvement; hard-normal regularization is an auxiliary constraint; decoupled fusion mainly affects instance ranking.
Förderung
- Pioneer and Leading Goose R&D Program of Zhejiang Province
Zitieren
APA
Wang, Q., Shi, L., He, F., Xiao, Z., Wang, X., Song, F., Yuan, T., Zhan, K., & Zhou, J. (2026). AsymLoc: Asymmetric Point Supervision for Cross-Dataset Zero-Shot 3D Anomaly Localization. Sensors, 26(22), 7109. https://doi.org/10.3390/s26227109
Vancouver
Wang Q, Shi L, He F, Xiao Z, Wang X, Song F, et al. AsymLoc: Asymmetric Point Supervision for Cross-Dataset Zero-Shot 3D Anomaly Localization. Sensors. 2026;26(22):7109. doi:10.3390/s26227109
BibTeX
@article{Wang2026,
author = {Wang, Qianhui and Shi, Leilei and He, Feng and Xiao, Zhongkai and Wang, Xinyun and Song, Fan and Yuan, Tianshun and Zhan, Kaicheng and Zhou, Jiali},
title = {AsymLoc: Asymmetric Point Supervision for Cross-Dataset Zero-Shot 3D Anomaly Localization},
journal = {Sensors},
year = {2026},
volume = {26},
number = {22},
pages = {7109},
doi = {10.3390/s26227109}
}
EUVIMED