Supervised Learning of Edges and Object Boundaries

By Piotr Doll'ar, Zhuowen Tu, Serge J. Belongie. 2006

505 citations recorded by the source corpus.

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

EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning (2019, background): “There is a large body of literature that addresses “Image-to-Edges” problems [5, 10, 26, 29].”
Holistically-Nested Edge Detection (2015, methodology): “…theory on top of features arrived at through careful manual design, such as Statistical Edges [20], Pb [26], and gPb [1]; and III: heavily learning-based methods that are still reliant on features of human design, such as BEL [5], Multi-scale [28], Sketch Tokens [22], and Structured Forests [6].”
Fast Edge Detection Using Structured Forests (2014, methodology): “In comparison to other learning-based approaches to edge detection, we considerably outperform BEL [13] which computes edges independently at each pixel given its surrounding image patch.”
Unsupervised Learning of Edges (2015, background): “Since the work of [10], which formulated edge detection as a binary classification problem, progressively more powerful learning paradigms have been employed, including multi-class classification [23], feature learning [30], regression [35], structured prediction [11], and deep learning [20, 5, 44].”
Pushing the Boundaries of Boundary Detection using Deep Learning (2015, background): “As in all works following the introduction of human-annotated datasets (Konishi et al., 2003; Martin et al., 2004), e.g. (Dollar et al., 2006; Arbelaez et al., 2011; Ren, 2008; Kokkinos, 2010a; Ren & Bo, 2012; Doll ´ ar & Zitnick, 2015), we use machine learning to optimize the performance of our…”
Crisp Boundary Detection Using Pointwise Mutual Information (2014, neutral): “[16] use supervised learning, along with a large number of features and multiple scales to learn edge prediction.”
Convolutional Oriented Boundaries: From Image Segmentation to High-Level Tasks (2016, methodology): “The BEL algorithm [27] attempts to learn an edge classifier in the form of a probabilistic boosting tree.”
Four-Chamber Heart Modeling and Automatic Segmentation for 3-D Cardiac CT Volumes Using Marginal Space Learning and Steerable Features (2008, unknown): “strated to have better performance on 2-D images [35]–[37]”
Constrained parametric min-cuts for automatic object segmentation (2010, methodology): “Learning methods have been applied for segmentation, to distinguish real from apparent contours [13, 10, 16], similar from dissimilar superpixels [15], and good from bad regions [25].”
Using contours to detect and localize junctions in natural images (2008, background): “…a single multiscale oriented signal: mPb(x, y, θ) = 9 ∑ i=1 αi · Gi(x, y, θ) (2) In order to introduce global information, we consider the first k + 1 generalized eigenvectors of the system (1), noted {v0, ...,vk}, where the corresponding eigenvalues {λ0, ..., λk} are such that 0 = λ0 ≤ ... ≤ λk .”
Computational analysis and learning for a biologically motivated model of boundary detection (2008, methodology): “in [16,54,76,77] were introduced in our system its ability to learn its recurrent connection weights from ground truth data would allow our system to surpass the performance of [16,54]; this is a direction that we intend to pursue in future research.”
Machine Learning for Aerial Image Labeling (2013, methodology): “For this reason recent work has relied on efficiently computable features in order to scale up to large context sizes [Kluckner and Bischof, 2009, Nguyen et al., 2007, Dollar et al., 2006].”
Richer Convolutional Features for Edge Detection (2016, background): “Later, researchers tended to manually design features using low-level cues such as intensity, gradient, and texture, and then employ sophisticated learning paradigm to classify edge and non-edge pixels [13, 44].”
Boundary Learning by Optimization with Topological Constraints (2010, methodology): “Boundary labelings from gPb [8] and Boosted Edge Learning (BEL) [3] are also displayed; additional comparisons are shown in the supplementarymaterial.”
Discriminative Sparse Image Models for Class-Specific Edge Detection and Image Interpretation (2008, methodology): “Precision-recall curves are presented on Figure 4 and are compared with Pb [18], BEL [5], UCM [2] and the recent gPb [16], which was published after that this paper was accepted.”
From contours to regions: An empirical evaluation (2009, background): “There have been significant advances in contour detection in the last few years [16] [37] [27] [12] [48] [22] [36].”
Semantic Image Segmentation with Contextual Hierarchical Models (2015, methodology): “methods, gPb-OWT-UCM (global probability of boundary followed by the oriented watershed transform and ultrametric contour maps) [10] and boosted edge learning (BEL) [55].”
Learning to localize detected objects (2012, background): “Many researchers have studied the problem of identifying edges that correspond to a particular object category [17, 7].”