Efficient Graph-Based Image Segmentationを用いた圃場図自動作成手法の検討

By 泰明岡田, 智一吉田, 智晴長尾. 2014

Subfields: Computer Vision and Pattern Recognition, Graph Learning

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

In Graph Learning

Citations in the corpus

EDTER: Edge Detection with Transformer (2022, methodology): “We compare our model with traditional detectors including Canny [6], Felz-Hutt [18], gPb-owt-ucm [2], SCG [64], Sketch Tokens [34], PMI [25], SE [15], OEF [21] and MES [50], and deep-learning-based detectors including DeepEdge [3], CSCNN [24], DeepCon-tour [48], HFL [4], HED [65], Deep Boundary [29]”
Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding (2020, neutral): “, [19]) to the input mesh.”
Multiscale Combinatorial Grouping for Image Segmentation and Object Proposal Generation (2015, background): “MCG is efﬁcient, its good generalization power makes it parameter free in practice, and it provides a ranked set of proposals that are competitive in all regimes of number of proposals.”
Piecewise Flat Embedding for Image Segmentation (2018, methodology): “1) For hierarchical clustering, we use finest level superpixel from [36], representing each superpixel with its average RGB values, and agglomeratively merging them until 2 clusters are left.”
Local Alignments for Fine-Grained Categorization (2014, methodology): “To generate the artificial noise we first decompose an image into a large number of superpixels using (Felzenszwalb and Huttenlocher, 2004).”
Unsupervised Multi-class Joint Image Segmentation (2014, background): “However, such techniques do not always obtain satisfactory results [4] since different parts of the same object may exhibit heterogeneous appearance.”
Salient Object Detection: A Discriminative Regional Feature Integration Approach (2013, methodology): “The initial way of constructing the region-level ground truth is to build multi-level segmentation by directly using the graph-based image segmentation approach (Felzenszwalb and Huttenlocher 2004), which is the same with the testing stage.”
Bayesian Semantic Instance Segmentation in Open Set World (2018, background): “Unsupervised segmentation: In contrast to learning based segmentation, unsupervised segmentation methods [23,24] are able to discover unknown objects without the strong supervisory training signal mentioned above.”
PiGraphs (2016, methodology): “The segmentations were manually labeled by grouping sets of segments obtained from an unsupervised normal-based segmentation [Felzenszwalb and Huttenlocher 2004], and assigning an object and part label to each group.”
Object discovery in 3D scenes via shape analysis (2013, methodology): “To partition a scene mesh into segments, we treat the mesh as a graph and use a graphbased segmentation algorithm proposed by Felzenszwalb and Huttenlocher [25].”
Learning Deep Structured Multi-Scale Features using Attention-Gated CRFs for Contour Prediction (2017, methodology): “Felz-Hut [11], MeanShift [7], Normalized Cuts [32], ISCRA [30], gPb-ucm [1], SketchTokens [22], MCG [28], LEP [47], and more recent CNN-based methods, including DeepEdge [2], DeepContour [31], HED [38], CEDN [42], COB [24].”
Learning Integral Objects With Intra-Class Discriminator for Weakly-Supervised Semantic Segmentation (2020, methodology): “To this end, we first refine the ICD scores Si,k,c by averaging their values in each superpixel [10], resulting in refined ICD score S′ i,k,c.”
Revisiting active perception (2016, background): “Representatives of such algorithms can be found in Felzenszwalb and Huttenlocher (2004), Alpert et al. (2007).”
SLIC Superpixels Compared to State-of-the-Art Superpixel Methods (2012, methodology): “GS04 – Felzenszwalb and Huttenlocher [8] propose an alternative graph-based approach that has been applied to generate superpixels.”
Spectral Clustering with Graph Neural Networks for Graph Pooling (2019, methodology): “Image Segmentation Given an image, we build a region adjacency graph (Trémeau & Colantoni, 2000) using as nodes the regions generated by an over-segmentation procedure (Felzenszwalb & Huttenlocher, 2004).”
Learning of perceptual grouping for object segmentation on RGB-D data (2014, methodology): “We employ graph-cut segmentation on the graph, introduced by Felzenszwalb and Huttenlocher [17], using the probability values as the pairwise energy terms to find a global optimum for object segmentation.”
Unknown Object Segmentation from Stereo Images (2021, background): “Unknown Object Instance Segmentation: Early works in UOIS mainly build on low-level image features based on boundaries, connectivity or symmetry [30], [31] to segment unknown instances.”