This paper introduces Rough Path Signature-Guided Geometry Augmentation (RPS-GA), a novel method to improve few-shot industrial defect detection. RPS-GA treats Canny edge contours as ordered paths and uses their signature responses to create a spatial map highlighting boundary structures. When tested with an unmodified YOLOv8n detector on the NEU-DET and PCB-Defect datasets, RPS-GA significantly boosted performance, especially with limited labeled data. AI
IMPACT This research offers a new technique to improve few-shot learning for industrial defect detection, potentially reducing the need for large labeled datasets in manufacturing.
RANK_REASON The cluster contains a research paper detailing a new methodology for a specific AI task. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Hugging Face Daily Papers →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
