A Statistical Approach to Texture Classification from Single Images

By Manik Varma, Andrew Zisserman. 2005

481 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

Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations (2016, methodology): “Next, textures were used to study objects (Varma and Zisserman 2005), while other methods learned to predict colors (Ferrari and Zisserman 2007).”
Describing Textures in the Wild (2013, methodology): “The first one is the use of the SIFT descriptors, originally developed for object matching [20], that are more distinctive that local descriptors popular in texture analysis such as filter banks [13, 19, 36], local intensity patterns [23], and patches [35].”
Hipster Wars: Discovering Elements of Fashion Styles (2014, methodology): “In particular, we calculate a vector of the following features at each pixel within a patch centered around the pixel: a) RGB color value, b) Lab color value, c) MR8 texture response [28] (to encode local patterns) d) HOG descriptor [5] (to measure local object shape), e) Distance from image border,”
Exploring features in a Bayesian framework for material recognition (2010, methodology): “A variety of methods based on texton representations [6, 18, 29 ], bidirectional histograms [8] and image patches [30] have been successful at classifying CURET surfaces (> 95% accuracy).”
TextonBoost for Image Understanding: Multi-Class Object Recognition and Segmentation by Jointly Modeling Texture, Layout, and Context (2007, methodology): “For this we use textons (Malik et al. 2001) which have been proven effective in categorizing materials ( Varma and Zisserman 2005 )a s well as generic object classes (Winn et al. 2005).”
From BoW to CNN: Two Decades of Texture Representation for Texture Classification (2018, methodology): “Later, Hay-man et al. [82] showed that SVM could further enhance the texture classiﬁcation performance of MR8 features, giving a 98 .”
Runway to Realway: Visual Analysis of Fashion (2015, background): “The first is a histogram of MR8 responses [23] quantized into 256 visual words.”
Learning to Recognize Daily Actions Using Gaze (2012, methodology): “We compute texture descriptors using the method of [25] and quantize them to 256 kmeans centers.”
Rotation Invariant Image Description with Local Binary Pattern Histogram Fourier Features (2009, methodology): “, [2], [3], [5].”
Extended local binary patterns for texture classification (2012, methodology): “All the results are as originally reported, except for those of VZ-MR8 and VZ-Joint, which are obtained by us using the exact same experimental setup as Varma and Zisserman did [2, 3].”
SpArSe: Sparse Architecture Search for CNNs on Resource-Constrained Microcontrollers (2019, background): “4 Results We report results on a variety of datasets: MNIST (55e3, 5e3, 10e3) [43], CIFAR10 (45e3, 5e3, 10e3) [39], CUReT (3704, 500, 1408) [60], and Chars4k (3897, 500, 1886) [25], corresponding to classification problems with 10, 10, 61, and 62 classes, respectively, with the training/validation/t”
Recognizing Materials Using Perceptually Inspired Features (2013, background): “2008; Romeiro and Zickler 2010) and the recognition of 3-D textures (Dana et al. 1999; Dana and Nayar 1998; Cula and Dana 2004a; Nillius and Eklundh 2004; Caputo et al. 2005; Varma and Zisserman 2005; Caputo et al. 2007; Varma and Zisserman 2009).”