Uni-3DAD: Gan-inversion aided universal 3D anomaly detection on model-free products

Jiayu Liu, Shancong Mou, Nathan Gaw, Yinan Wang

Abstract: Highlights•Our proposed method detects defects on those industrial products without the design file.•It fuses feature- and reconstruction-based methods for detecting all defect types.•It maintains an unsupervised nature and outperforms the state-of-the-art methods.