CVSep 14, 2015
gSLICr: SLIC superpixels at over 250Hz
arXiv:1509.04232v177 citationsHas Code
AI Analysis
This enables faster superpixel segmentation for computer vision applications, though it is an incremental improvement through parallelization.
The authors tackled the problem of slow superpixel segmentation by developing a parallel GPU implementation of SLIC, achieving speedups of up to 83× compared to the standard sequential version.
We introduce a parallel GPU implementation of the Simple Linear Iterative Clustering (SLIC) superpixel segmentation. Using a single graphic card, our implementation achieves speedups of up to $83\times$ from the standard sequential implementation. Our implementation is fully compatible with the standard sequential implementation and the software is now available online and is open source.