Geometric blur for template matching

By Alexander C. Berg, J. Malik. Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001, 2001

344 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/6cd135bf1fafd67ca284f3168ee88269789d8f9c

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

Observing Human-Object Interactions: Using Spatial and Functional Compatibility for Recognition (2009, background): “With the same neurons involved in execution and perception, a link between object recognition and action understanding has been established [38] in humans.”
Names and faces in the news (2004, methodology): “We use the geometric blur of [5] applied to grayscale patches as the features for our SVM.”
The Cross-Depiction Problem: Computer Vision Algorithms for Recognising Objects in Artwork and in Photographs (2015, methodology): “For BoW and FVs: for each descriptor we built an SVM classifier using a χ2 kernel.”
Object detection using a max-margin Hough transform (2009, methodology): “Weights are learned on codebooks generated using kmeans clustering of Geometric Blur (GB) features [3] sampled uniformly along the edges in an image.”
Matching with shape contexts (2000, positive): “Extension to the Continuous Case Berg and Malik [6] developed a descriptor which is akin to a shape context for grayscale images.”
Picking the best DAISY (2009, background): “a, the polar Gaussian pooling approach which has origins in geometric blur [2]) is one of the best for designing discriminative local image descriptors [23]; second, Tola et al.”
Gaussian Processes for Object Categorization (2010, methodology): “To describe each region or patch, we can choose from an array of descriptors designed to capture local texture while maintaining some invariance to small shifts and rotations, such as SIFT (Lowe 2004), shape context (Belongie et al. 2001), or geometric blur (Berg and Malik 2001).”
Ieee Transactions on Pattern Analysis and Machine Intelligence 1 Large Displacement Optical Flow: Descriptor Matching in Variational Motion Estimation (2011, neutral): “We investigate here three different methods: one based on region matching as proposed in [10], one based on HOG descriptors [15], and one based on geometric blur (GB) [5].”
Learning Globally-Consistent Local Distance Functions for Shape-Based Image Retrieval and Classification (2007, methodology): “Some of the best results on Caltech101[25, 6] have been achieved using geometric blur features[2].”
Animals on the Web (2006, background): “The geometric blur descriptor [5] first produces sparse channels from the gray scale image, in this case, half-wave rectified oriented edge filter responses at three orientations yielding six channels.”
Learning The Discriminative Power-Invariance Trade-Off (2007, background): “For the ﬁrst descriptor, GB, the distance between two images is f GB F 1 i and F 2 j are Geometric Blur features in the two images.”
Recent advances and trends in visual tracking: A review (2011, methodology): “Kembhavi et al. [64] proposed an Incremental Multiple Kernel Learning (IMKL) approach for object recognition, which combines the Pyramidal Histogram of Oriented Gradients (PHOG) [65] and Geometric Blur [66] together.”
Recognition using regions (2009, methodology): “To exploit multiple image cues, we extract four types of region descriptors (two types of shape, color and texture, all described in Section 2.2), as well as one point descriptor (Geometric Blur or GB [ 4 ]).”
Finding iconic images (2009, background): “Geometric Blur Features: are shape descriptors [2] that have been shown to perform quite well on object recognition tasks [24, 25].”
SVM-KNN: Discriminative Nearest Neighbor Classification for Visual Category Recognition (2006, background): “[26]), or the SIFT descriptor([24]), or the geometric blur descriptor([4]).”
Gabor-Filtering-Based Completed Local Binary Patterns for Land-Use Scene Classification (2015, methodology): “In [11], four types of features consist of DAISY [12], geometric blur [13], SIFT [10], and self-similarity [14] were used within the framework of multifeature joint sparse coding with spatial relation constraint.”