SLIC Superpixels ?

By Radhakrishna Achanta, Appu Shaji, Kevin Smith, Aurélien Lucchi, Pascal Fua, Sabine Süsstrunk. 2010

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Understanding Convolution for Semantic Segmentation (2017, methodology): “We then use the SLIC algorithm (50000 superpixels per example, color ratioM = 0 )[1] to gather the superpixels of the example, and merge nearby superpixels that have the same label.”
Are spatial and global constraints really necessary for segmentation? (2011, methodology): “To do so, we over-segment the image into superpixels using the SLIC algorithm [23].”
Robust Superpixel Tracking (2014, methodology): “In particular, superpixels have been applied to image segmentation and object recognition with demonstrated success [20], [23], [24], [27], [28].”
Saliency Detection via Absorbing Markov Chain (2013, unknown): “We construct a single layer graph G(V,E) with superpixels [3] as nodes V and the links between pairs of nodes as edges E.”
Salient object detection via bootstrap learning (2015, methodology): “In this paper, we compute a fixed number of superpixels from an input image using the Simple Linear Iterative Clustering (SLIC) method [2].”
Consistent Video Saliency Using Local Gradient Flow Optimization and Global Refinement (2015, methodology): “We ﬁrst apply the SLIC [25] to abstract each frame We adopt the large displacement motion estimation algorithm [39] to compute the optical ﬂow.”
Saliency detection via Cellular Automata (2015, methodology): “ed into N small superpixels by the simple linear iterative clustering (SLIC) algorithm [2].”
Range sensor and silhouette fusion for high-quality 3D Scanning (2015, methodology): “The interface, with a sample object, is shown in Figure 4(a); users select computed SLIC superpixels (marked with yellow boundaries) belonging to the object.”
Deep Contrast Learning for Salient Object Detection (2016, methodology): “In this paper, we use a slightly modified version of the SLIC algorithm [2], which uses geodesic image distance [9] during K-means clustering in the CIELab color space.”
Saliency Detection via Graph-Based Manifold Ranking (2013, methodology): “In this work, each node is a superpixel generated by the SLIC algorithm [3].”
An Exemplar-Based CRF for Multi-instance Object Segmentation (2014, methodology): “We use the SLIC method [1] to compute a superpixel representation of the test images tuned to yield about 1500 superpixels per image.”
Multi-focus Attention Network for Efficient Deep Reinforcement Learning (2017, methodology): “We believe that we can apply more sophisticated methods like super-pixel segmentation (Achanta et al. 2010) or spatial transformer networks (Jaderberg, Simonyan, and Zisserman 2015).”
Superpixels: An evaluation of the state-of-the-art (2016, methodology): “From the ranking in Table 3, we recommend 6 algorithms for use in practice, thereby covering a wide range of application scenarios: ETPS [84], SEEDS [80], ERS 1075 [47], CRS [78, 79], ERGC [68] and SLIC [74].”
Semantic Object Parsing with Graph LSTM (2016, methodology): “We use the SLIC over-segmentation method [39] to generate averagely 1,000 superpixels for each image.”
Deep Neural Networks Segment Neuronal Membranes in Electron Microscopy Images (2012, methodology): “The former approach uses haar-like features and texture histograms computed on a small region around the pixel of interest, whereas the latter uses sophisticated rotational [17] and ray [34] features computed on superpixels [3].”
What Makes a Patch Distinct? (2013, methodology): “The first step is solved by using the SLIC superpixels [2], already computed in Section 2.”
Depth Matters: Influence of Depth Cues on Visual Saliency (2012, methodology): “Next, we over-segment the color image into superpixels [20].”