GuidedFlow: An Attention-Guided Framework for Anomaly Detection in Additive Manufacturing
2026-08-24 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionMachine Learning
AI summaryⓘ
The authors focus on improving quality control in 3D printing by detecting tiny defects using images and videos, which is tricky with current methods. They introduce GuidedFlow, a new model that uses attention mechanisms and a pre-trained neural network to better spot and locate anomalies over time and at different scales. Their tests on real 3D printed objects and a standard industrial dataset show GuidedFlow generally works better than existing state-of-the-art models in finding defects. This model helps address challenges in recognizing subtle problems especially when data is limited.
Additive ManufacturingAnomaly DetectionNormalizing FlowsAttention MechanismResNetSpatio-Temporal ModelingAUROCIndustrial Quality Control3D Printing DefectsSmall-Data Learning
Authors
Sosmita Paul, Krishna Roy
Abstract
Additive Manufacturing (AM) plays a vital role in the ongoing industrial revolution. However, quality control remains crucial and challenging due to printing defects or potential cyber-physical intrusions. Image or video-based anomaly detection is a key effort towards addressing these challenges. Various approaches have been explored in this domain, including reconstruction-based, embedding-based, and flow-based methods. Though normalizing flow-based methods address some of the core challenges of unforeseen defects and generalization while maintaining detection performance, existing approaches struggle with tiny/stringing defects common in 3D printing. In a small-data setting, this poses a limitation in generalization. To address these limitations, we propose \textbf{GuidedFlow}, a novel attention-guided normalizing flow model for anomaly detection and localization. GuidedFlow employs a pre-trained ResNet model, fine-tuned on the domain dataset. An attention-guided spatial and temporal flow framework models the dynamics across multiple scales and frames. A Spatio-Temporal Attention Network (SAN) enables the flow model to prioritize relevant contextual cues from input frames. We evaluate GuidedFlow on our AM3D-AD dataset, consisting of benign and anomalous real 3D printed object images and videos. We also conduct a comparative study using the MVTec-AD industrial image anomaly detection dataset. Experimental results demonstrate that GuidedFlow outperforms most of the state-of-the-art models with enhanced detection accuracy and AUROC.