3D-PNAS generates 3D surface anomalies by sampling Perlin noise on a PCA-projected point cloud and displacing points along estimated normals, with parameters controlling scale, strength, and detail.
Anomalydiffusion: Few-shot anomaly image generation with diffusion model,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.GR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
3D-PNAS: 3D Industrial Surface Anomaly Synthesis with Perlin Noise
3D-PNAS generates 3D surface anomalies by sampling Perlin noise on a PCA-projected point cloud and displacing points along estimated normals, with parameters controlling scale, strength, and detail.