gSLICr: SLIC superpixels at over 250Hz

Abstract

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.

Cited by in corpus (10)

• Co-Fusion: Real-time Segmentation, Tracking and Fusion of Multiple Objects
• Video Propagation Networks
• Efficient semantic image segmentation with superpixel pooling
• FeatureLego: Volume Exploration Using Exhaustive Clustering of Super-Voxels
• Learning by Analogy: Reliable Supervision from Transformations for Unsupervised Optical Flow Estimation
• Recognizing Image Objects by Relational Analysis Using Heterogeneous Superpixels and Deep Convolutional Features
• Super-BPD: Super Boundary-to-Pixel Direction for Fast Image Segmentation
• Superpixel Segmentation Using Gaussian Mixture Model
• Points2Polygons: Context-Based Segmentation from Weak Labels Using Adversarial Networks
• Superpixel Sampling Networks