Sparse Coding on Local Spatial-Temporal Volumes for Human Action Recognition

Zhu, Yada; Zhao, Xu; Fu, Yun; Liu, Yuncai

doi:10.1007/978-3-642-19309-5_51

Cited by 65 publications

(62 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Aiming to alleviate the quantization errors in the BoW model, sparse coding has also been introduced to action recognition to learn more compact and richer representations of human actions [48,49,12].…”

Section: Sparse Codingmentioning

confidence: 99%

“…In order to obtain a more accurate and discriminative representation, an approach by encoding local 3D spatial-temporal gradient features was proposed by Zhu et al [49] in which the sparse coding framework is used for the final action representation. A local spatial-temporal feature is transformed to a linear combination of a few atoms in a trained dictionary.…”

Section: Sparse Codingmentioning

confidence: 99%

“…2. The results in [48] and [49] which use sparse coding on this dataset are 85,73% and 94.92%. A different experimental setting, i.e., using 599 video clips in total, is employed in [49] for validation.…”

Section: On the Kth Datasetmentioning

confidence: 99%

See 2 more Smart Citations

Action recognition via spatio-temporal local features: A comprehensive study

Zhen

Shao

2016

Image and Vision Computing

View full text Add to dashboard Cite

Local methods based on spatio-temporal interest points (STIPs) have shown their effectiveness for human action recognition. The bag-of-words (BoW) model has been widely used and dominated in this field. Recently, a large number of techniques based on local features including improved variants of the BoW model, sparse coding (SC), Fisher kernels (FK), vector of locally aggregated descriptors (VLAD) as well as the naive Bayes nearest neighbor (NBNN) classifier have been proposed and developed for visual recognition. However, some of them are proposed in the image domain and have not yet been applied to the video domain and it is still unclear how effectively these techniques would perform on action recognition. In this paper, we provide a comprehensive study on these local methods for human action recognition. We implement these techniques and conduct comparison under unified experimental settings on three widely used benchmarks, i.e., the KTH, UCF-YouTube and HMDB51 datasets. We discuss insightfully the findings from the experimental results and draw useful conclusions, which are expected to guide practical applications and future work for the action recognition community.

show abstract

Section: Sparse Codingmentioning

confidence: 99%

Section: Sparse Codingmentioning

confidence: 99%

See 1 more Smart Citation

Action recognition via spatio-temporal local features: A comprehensive study

Zhen

Shao

2016

Image and Vision Computing

View full text Add to dashboard Cite

show abstract

“…Recently, many researchers have given attention to dictionary learning [4]- [6]. However, traditional dictionary learning usually neglects the similarity among the coding coefficients and have poor performances in dealing with non-linearly separable data.…”

Section: Introductionmentioning

confidence: 99%

Learning a Similarity Constrained Discriminative Kernel Dictionary from Concatenated Low-Rank Features for Action Recognition

Huang

Wang

et al. 2016

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYTraditional low-rank feature lose the temporal information among action sequence. To obtain the temporal information, we split an action video into multiple action subsequences and concatenate all the low-rank features of subsequences according to their time order. Then we recognize actions by learning a novel dictionary model from concatenated low-rank features. However, traditional dictionary learning models usually neglect the similarity among the coding coefficients and have bad performance in dealing with non-linearly separable data. To overcome these shortcomings, we present a novel similarity constrained discriminative kernel dictionary learning for action recognition. The effectiveness of the proposed method is verified on three benchmarks, and the experimental results show the promising results of our method for action recognition.

show abstract

“…Recently, it has become more and more popular to employ sparse representation-based methods for various computer vision tasks, such as image classification [1], face recognition [2], [3], and human action recognition [4]- [8]. Sparse representation-based methods for human action recognition first compute the sparse feature representation with learned dictionary and then pool over the entire video to form the final representation, where average pooling and max pooling are usually used.…”

Section: Introductionmentioning

confidence: 99%

Statistics on Temporal Changes of Sparse Coding Coefficients in Spatial Pyramids for Human Action Recognition

Wang

et al. 2015

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYTraditional sparse representation-based methods for human action recognition usually pool over the entire video to form the final feature representation, neglecting any spatio-temporal information of features. To employ spatio-temporal information, we present a novel histogram representation obtained by statistics on temporal changes of sparse coding coefficients frame by frame in the spatial pyramids constructed from videos. The histograms are further fed into a support vector machine with a spatial pyramid matching kernel for final action classification. We validate our method on two benchmarks, KTH and UCF Sports, and experiment results show the effectiveness of our method in human action recognition.

show abstract

Sparse Coding on Local Spatial-Temporal Volumes for Human Action Recognition

Cited by 65 publications

References 22 publications

Action recognition via spatio-temporal local features: A comprehensive study

Action recognition via spatio-temporal local features: A comprehensive study

Learning a Similarity Constrained Discriminative Kernel Dictionary from Concatenated Low-Rank Features for Action Recognition

Statistics on Temporal Changes of Sparse Coding Coefficients in Spatial Pyramids for Human Action Recognition

Contact Info

Product

Resources

About