GenAU augments a vision-language model with segmentation tokens to unify image-level anomaly detection, pixel-level segmentation, multi-type classification, and language-based defect analysis in a single instruction-following architecture.
Patel, and Isht Dwivedi
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
A new method combines noisy point generation, multi-scale feature extraction, and implicit surface discrimination to learn signed distance functions that detect anomalies in 3D point clouds, reporting AUROC gains of 2.1% and 3.6% on two benchmarks.
citing papers explorer
-
GenAU: Language-Grounded Industrial Anomaly Understanding with Vision-Language Models
GenAU augments a vision-language model with segmentation tokens to unify image-level anomaly detection, pixel-level segmentation, multi-type classification, and language-based defect analysis in a single instruction-following architecture.
-
Learning Discriminative Signed Distance Functions from Multi-scale Level-of-detail Features for 3D Anomaly Detection
A new method combines noisy point generation, multi-scale feature extraction, and implicit surface discrimination to learn signed distance functions that detect anomalies in 3D point clouds, reporting AUROC gains of 2.1% and 3.6% on two benchmarks.