Motion Interchange Patterns for Action Recognition in Unconstrained Videos

Abstract.A new method for action modelling is proposed, which combines the trajectory beam obtained by semi-dense point tracking and a local binary trend description inspired from the Local Binary Patterns (LBP). The semi dense trajectory approach represents a good trade-off between reliability and density of the motion field, whereas the LBP component allows to capture relevant elementary motion elements along each trajectory, which are encoded into mixed descriptors called Motion Trend Patterns (MTP). The combination of those two fast operators allows a real-time, on line computation of the action descriptors, composed of space-time blockwise histograms of MTP values, which are classified using a fast SVM classifier. An encoding scheme is proposed and compared with the state-of-the-art through an evaluation performed on two academic action video datasets.

show abstract

“…In our experiments, the average number is set to 5 000 for a good result. [18] 82.36 [12] 93.0 [19] 88.38 [9] 93.8 [10] 90.8 [17] 90.5 [6] 95.0…”

Section: Methodsmentioning

confidence: 99%

“…They capture the motion effect on the local structure of self-similarities considering 3 neighbourhood circles at different instants. Kliper-Gross et al extended this idea to Motion Interchange Patterns [12], which encodes local changes in different motion directions.…”

Section: Introductionmentioning

confidence: 99%

Motion Trend Patterns for Action Modelling and Recognition

Nguyễn

Manzanera

Garrigues

2013

Computer Analysis of Images and Patterns

View full text Add to dashboard Cite

show abstract

“…In particular, on the HMDB51 dataset, the improvement over the best reported result to date 1 is about 3%. [14] 93.9% Le et al [14] 75.8% Sadanand et al [21] 26.9% Ji et al [8] 90.2% B. et al [2] 76.5% Orit et al [10] 29.2% Wang et al [25] 95% Wang et al [25] 84.1% Wang et al [25] 46.6% Our Method 95.6% Our Method 86.56% Our Method 49.22%…”

Section: Spatio-temporal Context Descriptorsmentioning

confidence: 99%

Exploring Motion Boundary based Sampling and Spatial-Temporal Context Descriptors for Action Recognition

Peng¹,

Qiao²,

Peng³

et al. 2013

Procedings of the British Machine Vision Conference 2013

View full text Add to dashboard Cite

Feature representation is important for human action recognition. Recently, Wang et al. [25] proposed dense trajectory (DT) based features for action video representation and achieved state-of-the-art performance on several action datasets. In this paper, we improve the DT method in two folds. Firstly, we introduce a motion boundary based dense sampling strategy, which greatly reduces the number of valid trajectories while preserves the discriminative power. Secondly, we develop a set of new descriptors which describe the spatial-temporal context of motion trajectories. To evaluate the performance of the proposed methods, we conduct extensive experiments on three benchmarks including K-TH, YouTube and HMDB51. The results show that our sampling strategy significantly reduces the computational cost of point tracking without degrading performance. Meanwhile, we achieve superior performance than the state-of-the-art methods by utilizing our spatial-temporal context descriptors.

show abstract

“…Further, Wang et al [26] extend the dense sampling approach by tracking the interest points using a dense optical flow field. We have observed that trajec-tories obtained by a Kanade-Lucas-Tomasi (KLT) tracker [8], densely sampled [26] or one of the variants [6,7,9,27], have been consistently performing well on several benchmark action recognition datasets. Early work by Albright et al [2] indicates that is it sufficient to distinguish human actions by tracking the body joint positions.…”

Section: Introductionmentioning

confidence: 99%

“…In the first approach, a variant of trajectories [7,9] and/or new local feature descriptors [6] is proposed. In the second approach, feature histograms are computed in different volumes obtained my dividing the video along height, width and time, and then aggregated.…”

Section: Introductionmentioning

confidence: 99%

Ordered Trajectories for Large Scale Human Action Recognition

Murthy

Goecke

2013

2013 IEEE International Conference on Computer Vision Workshops

View full text Add to dashboard Cite

Recently, a video representation based on dense trajectories has been shown to outperform other human action recognition methods on several benchmark datasets. In dense trajectories, points are sampled at uniform intervals in space and time and then tracked using a dense optical flow field. The uniform sampling does not discriminate objects of interest from the background or other objects. Consequently, a lot of information is accumulated, which actually may not be useful. Sometimes, this unwanted information may bias the learning process if its content is much larger than the information of the principal object(s) of interest. This can especially escalate when more and more data is accumulated due to an increase in the number of action classes or the computation of dense trajectories at different scales in space and time, as in the Spatio-Temporal Pyramidal approach. In contrast, we propose a technique that selects only a few dense trajectories and then generates a new set of trajectories termed 'ordered trajectories'. We evaluate our technique on the complex benchmark HMDB51, UCF50 and UCF101 datasets containing 50 or more action classes and observe improved performance in terms of recognition rates and removal of background clutter at a lower computational cost.

show abstract

Motion Interchange Patterns for Action Recognition in Unconstrained Videos

Cited by 147 publications

References 32 publications

Motion Trend Patterns for Action Modelling and Recognition

Motion Trend Patterns for Action Modelling and Recognition

Exploring Motion Boundary based Sampling and Spatial-Temporal Context Descriptors for Action Recognition

Ordered Trajectories for Large Scale Human Action Recognition

Contact Info

Product

Resources

About