Classifying Images of Materials: Achieving Viewpoint and Illumination Independence

By Manik Varma, Andrew Zisserman. European Conference on Computer Vision, 2002

449 citations recorded by the source corpus.

Subfields: Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/7417d7dcf6152736612e3f04ccc72731dc8d9505

How researchers see this paper

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

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

Character Recognition in Natural Images (2009, methodology): “Maximum Response of filters (MR8) (Varma and Zisserman, 2002) is a texture descriptor based on a set of 38 filters but only 8 responses.”
A Statistical Approach to Texture Classification from Single Images (2005, result): “A preliminary version of these results appeared in Varma and Zisserman (2002) .”
Virtual Microscopy and Grid-enabled Decision Support for Large Scale Analysis of Imaged Pathology Specimens (2009, methodology): “4) The Maximum Response (MR) Filter Banks [20]: The MR filter bank is quite similar to the LM filter bank.”
Class-specific material categorisation (2005, methodology): “While the choice between four state-of-the-art descriptors [24, 25, 20, 16] has little effect, using Support Vector Machines (SVMs) over Nearest Neighbour classifiers proves beneficial, as it did for material identification in [12].”
Texture classification: are filter banks necessary? (2003, methodology): “Further details of the filter bank, as well as pre and post image processing steps, are given in [17].”
A Statistical Approach to Texture Classification from Single Images (2004, background): “Keywords: material classification, 3D textures, textons, filter banks, rotation invariance”
The Scale of Geometric Texture (2012, background): “Other approaches have involved explicitly training a classifier on multiple images of the texture from different viewpoints and illumination conditions [13, 22].”
Classifying materials in the real world (2010, methodology): “3) that using SVMs leads to a gain in performance compared to the method proposed by Varma and Zisserman [5,9].”
Context Encoding for Semantic Segmentation (2018, methodology): “For a given input image, hand-engineered features are densely extracted using SIFT [38] or ﬁlter bank responses [30,48].”
A comprehensive review of current local features for computer vision (2008, methodology): “19(c); and Varma and Zisserman [81,82] developed the maximum response ﬁlter bank based on the root ﬁlter set, as shown in Fig.”
A Dynamic Conditional Random Field Model for Joint Labeling of Object and Scene Classes (2008, background): “Further, we observe that normalization [26] as well as multi-scale features are helpful to improve the classification results.”
A Bayesian Model for Simultaneous Image Clustering, Annotation and Object Segmentation (2009, background): “Each superpixel is represented by both color and texture descriptors, based on the local RGB, hue [25] featur vectors and also the output of maximum response (MR) filter banks [22] ( http://www.”
Viewpoint Invariant Texture Description Using Fractal Analysis (2009, methodology): “The modified texton-based histogram [22,7 ,36] reduces the negative effects of changes in illumination, but in trade-off t o larger sensitivity to global geometric transforms.”