Sparselet Models for Efficient Multiclass Object Detection

By Hyun Oh Song, Stefan Zickler, Tim Althoff, Ross B. Girshick, Mario Fritz, Christopher Geyer, Pedro F. Felzenszwalb, Trevor Darrell. 2012

92 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Industrial and Manufacturing Engineering

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Industrial and Manufacturing Engineering

Citations in the corpus

Scalable Object Detection Using Deep Neural Networks (2013, background): “To address the latter issue, Song et al. [12] use a low-dimensional part basis, shared across all object classes.”
Deformable part models are convolutional neural networks (2014, methodology): “When implemented on a GPU, it is faster to loosely bound the pooling region and use a brute-force, but parallel-friendly maximization, as was done in [37].”
Speeding up Convolutional Neural Networks with Low Rank Expansions (2014, methodology): “One way to exploit this redundancy is to approximate the filter set by a linear combination of a smaller basis set of M filters [32, 35, 36].”
Learning to Share Latent Tasks for Action Recognition (2013, background): “In object recognition, a number of papers have been published on transferring visual knowledge between different object categories [23, 18, 5, 19, 24, 26].”
Multiview RGB-D Dataset for Object Instance Detection (2016, methodology): “Examples of using these models in the table top settings similar to ours include [16, 27].”
Strathprints Institutional Repository Shuhui and Ren, Jinchang (2015) Effective and Efficient Midlevel Visual Elements-oriented Land-use Classification Using Vhr Remote Sensing Images. Ieee Transactions on Geoscience and Remote Sensing, 53 (8). Pp. 4238-4249. Issn 0196-2892 Effective and Efficient M (unknown): “However, the method proposed by Song et al. [23] for learning sparselets and activation vectors is obviously brittle, in which sparselets and activation vectors were approximately obtained by using greedy algorithms such as orthogonal matching pursuit algorithm (OMP) [25] [26] [27] without fully exp”
Fast, Accurate Detection of 100,000 Object Classes on a Single Machine (2013, background): “Very recently, [13, 17] consider scaling category-level detection by learning a set of shared basis parts in the form of steerable filters or sparselets.”
A coarse-to-fine approach for fast deformable object detection (2011, background): “[28] represent the object parts of a deformable model as a sparse linear composition of a reduced set of basic parts shared among different categories.”
Exemplar based Deep Discriminative and Shareable Feature Learning for scene image classification (2015, background): “In [43, 44], object part filters at the middle level are shared to represent a large number of object categories for object detection.”
Feature representation for statistical-learning-based object detection: A review (2015, methodology): “Unlike Sparselets, the basis elements of shufflet part filters are shiftable.”
30Hz Object Detection with DPM V5 (2014, methodology): “Vanilla DPM [22] is a version of DPM that can harness the power of GPU to speed-up object detectors.”
Sparse Modeling for Image and Vision Processing (2014, unknown): “…and audio processing [Lewicki, 2002, Elad and Aharon, 2006, Mairal et al., 2009, Yang et al., 2010a] and computer vision [Raina et al., 2007, Yang et al., 2009, Zeiler et al., 2011, Mairal et al., 2012, Song et al., 2012, Castrodad and Sapiro, 2012, Elhamifar et al., 2012, Pokrass et al., 2013].”
Training Deformable Part Models with Decorrelated Features (2013, background): “Popular approaches include cascades [11], coarse-to-ﬁne processing [22], and sparse representations [23, 24].”
Histograms of Sparse Codes for Object Detection (2013, background): “Most approaches have explored model structure, either through nonparametric mixtures or exemplars [21, 10], compositional grammar structure [15], supervised correspondences [5, 2], and low-dimensional projections [29, 25].”
The Fastest Deformable Part Model for Object Detection (2014, background): “Acceleration of Multi-category DPM Quite a large number of recent works [23, 27, 15, 3, 17] were proposed to accelerate DPM for multi-category detection, e.g. simultaneous detection of 20 categories on Pascal VOC.”
Large-Scale Live Active Learning: Training Object Detectors with Crawled Data and Crowds (2011, background): “Even with a human in the loop, annotation precision varies when using Web interfaces and crowds, and so some research explores ways to automatically provide quality assurance [13, 14].”