Differentiable NMS via Sinkhorn Matching for End-to-End Fabric Defect Detection
This addresses fabric defect detection for industrial quality control, offering an incremental improvement by making NMS differentiable and reducing annotation costs.
The paper tackled the problem of non-differentiable non-maximum suppression and costly annotations in fabric defect detection by proposing a differentiable NMS framework using Sinkhorn matching, achieving significant performance improvements on the Tianchi dataset while maintaining real-time speeds.
Fabric defect detection confronts two fundamental challenges. First, conventional non-maximum suppression disrupts gradient flow, which hinders genuine end-to-end learning. Second, acquiring pixel-level annotations at industrial scale is prohibitively costly. Addressing these limitations, we propose a differentiable NMS framework for fabric defect detection that achieves superior localization precision through end-to-end optimization. We reformulate NMS as a differentiable bipartite matching problem solved through the Sinkhorn-Knopp algorithm, maintaining uninterrupted gradient flow throughout the network. This approach specifically targets the irregular morphologies and ambiguous boundaries of fabric defects by integrating proposal quality, feature similarity, and spatial relationships. Our entropy-constrained mask refinement mechanism further enhances localization precision through principled uncertainty modeling. Extensive experiments on the Tianchi fabric defect dataset demonstrate significant performance improvements over existing methods while maintaining real-time speeds suitable for industrial deployment. The framework exhibits remarkable adaptability across different architectures and generalizes effectively to general object detection tasks.