Published on Sun Aug 17 2014

Action Classification with Locality-constrained Linear Coding

Hossein Rahmani, Arif Mahmood, Du Huynh, Ajmal Mian

We propose an action classification algorithm which uses Locality-constrained Linear Coding (LLC) to capture discriminative information of human body movements. We justify the use of LLC for encoding the block descriptor by demonstrating its superiority over Sparse Coding.

0
0
0
Abstract

We propose an action classification algorithm which uses Locality-constrained Linear Coding (LLC) to capture discriminative information of human body variations in each spatiotemporal subsequence of a video sequence. Our proposed method divides the input video into equally spaced overlapping spatiotemporal subsequences, 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.

Tue May 10 2016
Computer Vision
Action Recognition in Video Using Sparse Coding and Relative Features
This work presents an approach to category-based action recognition in video using sparse coding techniques. The proposed approach includes two main contributions. i) A new method to handle intra-class variations by decomposing each video into a reduced set of representative atomic action acts or key-sequences. ii)
0
0
0
Mon Sep 02 2013
Computer Vision
A Study on Unsupervised Dictionary Learning and Feature Encoding for Action Classification
Sparse coding performs consistently better than the other encoding methods in large complex dataset. For small simple dataset (i.e., KTH) with less variation, all the encoding strategies perform competitively. The strength of sophisticated encoding approaches comes from their corresponding dictionaries.
0
0
0
Fri Feb 13 2015
Computer Vision
Long-short Term Motion Feature for Action Classification and Retrieval
We propose a method for representing motion information for video classification and retrieval. We improve upon local descriptor based methods that have been among the most popular and successful models for representing videos. The desired local descriptors need to satisfy two requirements.
0
0
0
Thu Aug 01 2013
Computer Vision
Sparse Dictionary-based Attributes for Action Recognition and Summarization
We present an approach for dictionary learning of action attributes via appearance information maximization. We unify the class distribution and appearance information into an objective function for learning a sparse dictionary. The objective function maximizes the mutual information between what has been learned and what remains to be learned.
0
0
0
Thu Oct 15 2015
Computer Vision
Beyond Spatial Pyramid Matching: Space-time Extended Descriptor for Action Recognition
We address the problem of generating video features for action recognition. Instead of only coding motion information, location information is used as part of the encoding step. This method is a much more effective and efficient location encoding method as compared to the fixed grid model.
0
0
0
Fri Aug 29 2014
Computer Vision
Temporal Extension of Scale Pyramid and Spatial Pyramid Matching for Action Recognition
This paper proposes to encode equivalent temporal characteristics in video representations for action recognition. The experimental results on four benchmark datasets,UCF50, HMDB51, Hollywood2 and Olympic Sports, support our approach and significantly outperform state-of-the-art methods.
0
0
0