Vlfeat: an open and portable library of computer vision algorithms

By Andrea Vedaldi, Brian Fulkerson. 2010

3,531 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

Deep Residual Learning for Image Recognition (2015, background): “Both of them are powerful shallow representations for image retrieval and classification [4, 47].”
Making Convolutional Networks Shift-Invariant Again (2019, background): “Shift-equivariance can be recovered if features are extracted densely, for example tex-tons (Leung & Malik, 2001), the Stationary Wavelet Trans-form (Fowler, 2005), and DenseSIFT (Vedaldi & Fulkerson, 2008).”
Framing Image Description as a Ranking Task: Data, Models and Evaluation Metrics (2013, methodology): “…image representations used in this paper are very basic, in that they rely only on three different kinds of low-level pixel-based perceptual features that capture color, texture (Varma & Zisserman, 2005) and shape information in the form of SIFT descriptors (Lowe, 2004; Vedaldi & Fulkerson, 2008).”
Learning to See by Moving (2015, methodology): “SIFT features are extracted using a square window of size 72 pixels and a stride of 8 pixels using the open source code from [35].”
Inverting Visual Representations with Convolutional Networks (2015, methodology): “For all three methods we use implementations from the VLFeat library [23] with the default settings.”
DeltaPhish: Detecting Phishing Webpages in Compromised Websites (2017, methodology): “For each cell, a 31-dimensional HOG descriptor is computed, in which each bin represents a quantized direction and its value corresponds to the magnitude of gradients in that direction (we refer the reader to [36, 34, 35] for further details).”
Determinantal Point Processes for Machine Learning (2012, methodology): “• SIFT (2 variants): The images are processed with the vlfeat toolbox to obtain sets of 128-dimensional SIFT descriptors [96, 150].”
Nonparametric kernel estimators for image classification (2012, methodology): “Our implementation uses the PHOW function of the VLFEAT package [30] for feature extraction.”
MatConvNet: Convolutional Neural Networks for MATLAB (2014, background): “VLFeat –”
Multi-view Convolutional Neural Networks for 3D Shape Recognition (2015, methodology): “The Fisher vector image descriptor is implemented using VLFeat [36].”
Understanding Image Representations by Measuring Their Equivariance and Equivalence (2014, background): “HOG implementations (Vedaldi and Fulkerson 2010) do in fact explicitly provide such per-”
Bilinear CNN Models for Fine-Grained Visual Recognition (2015, methodology): “Fisher vector with SIFT (FV-SIFT) We implemented a FV baseline using dense SIFT features [28] extracted using VLFEAT [35].”
Coherent Multi-sentence Video Description with Variable Level of Detail (2014, methodology): “We densely sample the points in the hands’ neighborhood, extract color Sift features [20] on 4 channels (RGB+grey) and quantize them in a codebook of size 4000.”
SUN Database: Exploring a Large Collection of Scene Categories (2014, methodology): “SIFT descriptors are computed using the VLFeat library [48] and we also use a histogram intersection kernel as in [25].”
Evaluation of output embeddings for fine-grained image classification (2014, methodology): “We extract 128-dim SIFT from regular grids at multiple scales, reduce them to 64-dim using PCA, build a visual vocabulary with 256 Gaussians [53] and finally reduce the FVs to 4,096.”
CloudCV: Large-Scale Distributed Computer Vision as a Cloud Service (2015, methodology): “• Narrow-Focus Libraries: A number of toolboxes provide specialized implementations for specific tasks, e.g. Camera Calibration Toolbox [24], Structure-fromMotion toolboxes [31, 46, 50], Visual Features Library [49], and deep learning frameworks such as Caffe [32], Theano [19, 21], Torch [16] etc. •”
N4-Fields: Neural Network Nearest Neighbor Fields for Image Transforms (2014, methodology): “After neural codes for all patches are computed, nearest-neighbors search is done by means of k - d trees provided as a part of VLFeat package [24].”
Kornia: an Open Source Differentiable Computer Vision Library for PyTorch (2019, background): “VLFeat was written in C for efﬁciency and compatibility, with interfaces in MATLAB.”