Max-margin additive classifiers for detection

By Subhransu Maji, Alexander C. Berg. 2009

189 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Quasi-Monte Carlo Feature Maps for Shift-Invariant Kernels (2014, background): “Maji and Berg [2009] suggested random fe atur s for the intersection kernel k(x, z) = ∑d i=1min(xi, zi).”
Random Feature Maps for Dot Product Kernels (2012, background): “Subsequently Li et al. (2010) generalized this to a larger class of group invariant kernels while still invoking Bochner’s theorem.”
Compact Bilinear Pooling (2015, methodology): “This was later extended for the intersection kernel, χ(2) kernel and the exponential χ(2) kernel [35, 25, 36].”
Orthogonal Random Features (2016, background): “2 Related Works Explicit nonlinear random feature maps have been constructed for many types of kernels, such as intersection kernels [15], generalized RBF kernels [22], skewed multiplicative histogram kernels [14], additive kernels [24], and polynomial kernels [11, 18].”
Label-Embedding for Image Classification (2015, neutral): “This includes work on patch encoding (see [8] for a recent comparison), on kernel-based methods [55] with a recent focus on explicit embeddings [35], [60], on dimensionality reduction [55] and on compression [26], [49], [61].”
High-dimensional signature compression for large-scale image classification (2011, background): “Maji and Berg [18] and Wang et al. [31] then proposed efficient algorithms to learn IKSVMs.”
Meta-class features for large-scale object categorization on a budget (2012, background): “The first category comprises techniques to approximate nonlinear kernel distances via explicit feature maps [22, 31]: for many popular kernels in computer vision, these methods provide analytical mappings to slightly higherdimensional feature spaces where inner products approximate the kernel distan”
Compositional Models for Video Event Detection: A Multiple Kernel Learning Latent Variable Approach (2013, background): “A body of work has aimed at providing efficient training and evaluation with kernelized classifiers via algorithmic optimizations or additive linear approximations [15, 6, 10].”
Semantic segmentation using regions and parts (2012, background): “Our region scoring strategy is tractable thanks to the low dimensionality of the representation and the use of additive kernels [26, 25, 34], which combine the expressive power of non-linear classifiers with efficiency close to linear ones, both in training and testing.”
Spherical Random Features for Polynomial Kernels (2015, background): “Explicit nonlinear random feature maps have also been proposed for other types of kernels, such as intersection kernels [7], generalized RBF kernels [8], skewed multiplicative histogram kernels [9], additive kernels [10], and semigroup kernels [11].”
Efficient additive kernels via explicit feature maps (2010, methodology): “The approximations work as well as the exact kernels and better than Maji and Berg [ 13 ]’s approximation in the low dimensional regime.”
A Multi-View Embedding Space for Modeling Internet Images, Tags, and Their Semantics (2012, methodology): “To handle large numbers of images and high-dimensional features, we propose a scalable approach based on the idea of approximate kernel maps (Maji and Berg, 2009; Perronnin et al., 2010; Rahimi and Recht, 2007; Vedaldi and Zisserman, 2010).”
Towards good practice in large-scale learning for image classification (2012, background): “An explicit mapping of the image descriptors to efficiently deal with non-linear kernels was proposed in [21, 25, 36].”
Good Practice in Large-Scale Learning for Image Classification (2014, background): “Maji and Berg [51] proposed mappings for the Intersection Kernel (IK).”
Fast and scalable polynomial kernels via explicit feature maps (2013, methodology): “Instead of approximating the kernel matrix, recent approaches [10, 12, 16, 20, 21, 24] approximate the kernels by explicitly mapping data into a relatively low-dimensional random feature space.”
Improving the Fisher Kernel for Large-Scale Image Classification (2010, background): “[14], Maji and Berg [15], Perronnin et al .”
Kernel Pooling for Convolutional Neural Networks (2017, methodology): “[26] for the histogram intersection kernel and Vedaldi et al.”