Action Classification with Locality-Constrained Linear Coding

By Hossein Rahmani, Arif Mahmood, Du Q. Huynh, Ajmal Saeed Mian. 2014

25 citations recorded by the source corpus.

We propose an action classification algorithm which uses Locality-constrained Linear Coding (LLC) to capture discriminative information of human body variations in each spatio-temporal subsequence of a video sequence. Our proposed method divides the input video into equally spaced overlapping spatio-temporal sub sequences, each of which is decomposed into blocks and then cells. We use the Histogram of Oriented Gradient (HOG3D) feature to encode the information in each cell. We justify the use of LLC for encoding the block descriptor by demonstrating its superiority over Sparse Coding (SC). Our sequence descriptor is obtained via a logistic regression classifier with L2 regularization. We evaluate and compare our algorithm with ten state-of-the-art algorithms on five benchmark datasets. Experimental results show that, on average, our algorithm gives better accuracy than these ten algorithms.

Subfields: Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning

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

In Representation & Self-Supervised Learning

Citations in the corpus

Deep Multimodal Feature Analysis for Action Recognition in RGB+D Videos (2016, neutral): “22% HOG3D-LLC [86] 98.”
RGB-D-based action recognition datasets: A survey (2016, methodology): “UWA3D Multiview UWA3D Multiview Activity Dataset [47,48,51] (http://staff home.”
Histogram of Oriented Principal Components for Cross-View Action Recognition (2014, background): “HOPC is extracted at a point by projecting all three eigenvectors of the pointcloud within a local spatio-temporal support volume, scaled by their corresponding eigenvalues, on the vertices of a regular dodecahedron.”
Learning a Deep Model for Human Action Recognition from Novel Viewpoints (2016, unknown): “[8], [9], [10], [11], [12], [13], [14], [28], [44], [45], [46] deals”
A discussion on the validation tests employed to compare human action recognition methods using the MSR Action3D dataset (2014, unknown): “[56] 2014 - 90.”
Human Action Recognition From Various Data Modalities: A Review (2020, unknown): “Dataset Year Modality #Class #Subject #Sample #View KTH [347] 2004 RGB 6 25 2,391 1 Weizmann [40] 2005 RGB 10 9 90 1 IXMAS [348] 2006 RGB 11 10 330 5 HDM05 [349] 2007 RGB,S 130 5 2,337 1 Hollywood [350] 2008 RGB 8 - 430 Hollywood2 [351] 2009 RGB 12 - 3,669 MSR-Action3D [217] 2010 S,D 20 10 567 1 Oly”
Learning Action Recognition Model from Depth and Skeleton Videos (2017, background): “The depth sensor based human action recognition research can be broadly divided into three categories including skeleton data [6,12,25,33,34,38,40], depth images [15, 17,18,20,21,23,37,39] and depth skeleton [19,26,27,36] based methods.”
3D Action Recognition from Novel Viewpoints (2016, methodology): “Many methods [23, 31, 34, 36–38, 44, 47, 48, 55–57, 63, 67, 68, 74] have been proposed which achieve impressive action recognition results when videos are acquired from a common viewpoint.”
Learning a non-linear knowledge transfer model for cross-view action recognition (2015, background): “This is because the same action appears quite different when observed from dif ferent viewpoints [25-27].”