🤖 AI Summary
Existing superpixel methods struggle to achieve texture-aware pixel clustering under constrained computational budgets. To address this, we propose Nearest-Neighbor Superpixel Clustering (NNSC), a block-matching-based approach that departs from conventional pixel-wise K-means paradigms and—uniquely—introduces patch-level nearest-neighbor matching into superpixel generation, directly modeling local texture structure in the image patch space. NNSC integrates patch feature extraction, adaptive distance metric learning, and efficient graph-based clustering optimization, jointly preserving structural fidelity and ensuring computational efficiency. Evaluated on standard color and texture-rich benchmarks, NNSC achieves state-of-the-art performance: it significantly outperforms mainstream methods in segmentation accuracy—as measured by boundary recall (BE) and undersegmentation error (UE)—and runs 2–5× faster than existing texture-aware superpixel algorithms, establishing new benchmarks in both quality and speed.
📝 Abstract
Superpixels are widely used in computer vision applications. Nevertheless, decomposition methods may still fail to efficiently cluster image pixels according to their local texture. In this paper, we propose a new Nearest Neighbor-based Superpixel Clustering (NNSC) method to generate texture-aware superpixels in a limited computational time compared to previous approaches. We introduce a new clustering framework using patch-based nearest neighbor matching, while most existing methods are based on a pixel-wise K-means clustering. Therefore, we directly group pixels in the patch space enabling to capture texture information. We demonstrate the efficiency of our method with favorable comparison in terms of segmentation performances on both standard color and texture datasets. We also show the computational efficiency of NNSC compared to recent texture-aware superpixel methods.
