Action classification with locality-constrained linear coding

H. Rahmani, Arif Mahmood, Du Huynh, Ajmal Mian

Research output: Chapter in Book/Conference paperConference paperpeer-review

19 Citations (Scopus)


© 2014 IEEE. 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.
Original languageEnglish
Title of host publicationProceedings - International Conference on Pattern Recognition
Place of PublicationUSA
PublisherIEEE, Institute of Electrical and Electronics Engineers
ISBN (Print)9781479952083
Publication statusPublished - 2014
Event22nd International Conference on Pattern Recognition - Stockholm, Sweden
Duration: 24 Aug 201428 Aug 2014
Conference number: 109641


Conference22nd International Conference on Pattern Recognition
Abbreviated titleICPR 2014


Dive into the research topics of 'Action classification with locality-constrained linear coding'. Together they form a unique fingerprint.

Cite this