Large-scale image classification: Fast feature extraction and SVM training

By Yuanqing Lin, Fengjun Lv, Shenghuo Zhu, Ming Yang, Timothée Cour, Kai Yu, Liangliang Cao, Thomas S. Huang. 2011

481 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Optimization & Training

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

In Optimization & Training

Citations in the corpus

ImageNet Large Scale Visual Recognition Challenge (2014, methodology): “Im a g eN et L a rg e S ca le V isu a l R eco g n itio n C h a llen g e 1 7 ILSVRC 2010 Codename CLS LOC Insitutions Contributors and references Hminmax 54.4 Massachusetts Institute of Technology Jim Mutch, Sharat Chikkerur, Hristo Paskov, Ruslan Salakhutdinov, Stan Bileschi, Hueihan Jhuang IBM 70.1”
Deep Learning: Methods and Applications (2014, background): “Another set of popular studies on unsupervised deep feature learning for computer vision are based on deep sparse coding models [226].”
Towards Open World Recognition (2014, background): “Incremental Learning: As SVMs rose to prominence in for object recognition applications [46, 22], many incremental extensions to SVMs were proposed.”
The Computational Limits of Deep Learning (2020, methodology): “The NEC-UIUC team won the competition by using a novel algorithm that combines SIFT [125], LBP [126], two non-linear coding representations [127, 128], and stochastic SVM [129].”
Fisher vectors meet Neural Networks: A hybrid classification architecture (2015, methodology): “In 2010-2011, the leading approaches to ILSVRC [43, 31] were based on the Fisher Vector [38, 39] (FV) or the closely related Super Vector (SV) [57].”
Efficient Object Detection and Segmentation for Fine-Grained Recognition (2013, unknown): “[16].”
Studying Relationships between Human Gaze, Description, and Computer Vision (2013, background): “Recent advances have started to look at problems of recognition at a human scale, classifying or localizing thousands of object categories with reasonable accuracy [19, 24, 5, 6, 18].”
Deep learning for visual understanding (2017, unknown): “A remarkable result is shown in [141], in which the proposed averaging stochastic gradient descent (ASGD) scheme combined LCC and SVC algorithm to scale the image classi cation to large-scale dataset, and produced state-of-the-art results on ImageNet object recognition tasks prior to the rise of CNN”
Incremental Learning of NCM Forests for Large-Scale Image Classification (2014, background): “To address the problem of large data, there has been a wide use of online learning methods, such as stochastic gradient descent [13, 20].”
Large Scale Visual Recognition (2012, background): “One line of work is concerned with developing effective feature representations [95, 96, 97, 98] and achieving state of the art performances.”
Depth kernel descriptors for object recognition (2011, unknown): “Single layer sparse coding on top of SIFT features works surprisingly well [25], [6], and is the best-performing approach on the first ImageNet Large-scale Visual Recognition Challenge [18].”
Deep Convolutional Neural Networks for Image Classification: A Comprehensive Review (2017, background): “Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/neco_a_00990 by guest on 04 June 2021 Lin et al., 2011; Sánchez & Perronnin, 2011), as well as an approximate correlation between classification performance and network depth (see section 5.5.4).”
Learning to Search Efficiently in High Dimensions (2011, background): “Recent advances in computer vision show that it is desirable to represent images as numerical ve tors of as high as thousands or even millions of dimensions [12, 28].”
Hedging your bets: Optimizing accuracy-specificity trade-offs in large scale visual recognition (2012, background): “Our work is also inspired by an emerging trend in computer vision studying large scale visual recognition [19, 27, 13, 12, 31, 7, 26, 24].”
Evaluating knowledge transfer and zero-shot learning in a large-scale setting (2011, unknown): “5 Top 1 BoW[1] Sift LibLinear 1,000 80 91 BoW Sift MeanSGD 1,000 72 86 BoW+SPM rgSift MeanSGD 8,000 59 76 LLC+SPM rgSift MeanSGD 21,000 50 69 Fisher vector rgSift MeanSGD 32,768 43 61 LLC+SPM, Fisher rgSift MeanSGD 53,768 38 57 Fisher+SPM [3] Sift, Color SGD 262,144 34 LLC,SVC+SPM [2] Hog, Lbp ASGD ”
ImageNet Auto-Annotation with Segmentation Propagation (2014, background): “Several works tackle image classification [18,42,21,20,36, 17] or object detection in the fully supervised setting [17].”
Consistency versus Realizable H-Consistency for Multiclass Classification (2013, background): “(We note that this is a standard practice for large-scale multiclass image classification, see e.g. (Weston et al., 2010; Lin et al., 2011; Perronnin et al., 2012)).”
Metric Learning for Large Scale Image Classification: Generalizing to New Classes at Near-Zero Cost (2012, unknown): “The current state-of-the-art [8,10] uses efﬁ-cient linear SVM classiﬁers trained in a one-vs-rest manner in combination with high-dimensional bag-of-word [20] or Fisher vector representations [19].”
Face Spoof Detection With Image Distortion Analysis (2015, methodology): “Following the success of SVM [36] in signal processing [37], pattern recognition and classification applications [38], [39], we choose to use SVM via the LibSVM Library [40].”