Human Action Recognition by Semilatent Topic Models

Abstract

We propose two new models for human action recognition from video sequences using topic models. Video sequences are represented by a novel "bag-of-words" representation, where each frame corresponds to a "word." Our models differ from previous latent topic models for visual recognition in two major aspects: first of all, the latent topics in our models directly correspond to class labels; second, some of the latent variables in previous topic models become observed in our case. Our models have several advantages over other latent topic models used in visual recognition. First of all, the training is much easier due to the decoupling of the model parameters. Second, it alleviates the issue of how to choose the appropriate number of latent topics. Third, it achieves much better performance by utilizing the information provided by the class labels in the training set. We present action classification results on five different data sets. Our results are either comparable to, or significantly better than previously published results on these data sets.

Development of Biological Movement Recognition by Interaction between Active Basis Model and Fuzzy Optical Flow Division

“…Here, the experimental result of proposed approach is presented. As KTH human action database [39] has been used for benchmarking the accuracy of consistency with set of experiments used in [2, 22, 27, 43, 44], we made a set of our training map and test set for proposed technique on the entire dataset, in which the mixture of four scenarios videos was together. The dataset split into a set of training maps with five randomly selected subjects and a test part by residual subjects.…”

“…Using optical flow (as it is aforementioned) is one of the effective methods in human action recognition (see [27]). For having the features regarding motion of subject, layer-wise optical flow estimation has been done.…”

Comparative Study on Interaction of Form and Motion Processing Streams by Applying Two Different Classifiers in Mechanism for Recognition of Biological Movement

“…Similarity of the presented model has been deliberated in the assessment. KTH human action dataset is used for benchmarking and the evaluation assessment compared with state-of-the-art methods in the same dataset for consistency in the experimental results (see Figure 4) [11, 42–46]. Also it is noticeable, as previously mentioned, that the training map and action prototypes obtained from the random selection of the human action set in four different scenario videos from KTH and excluded from the testing set have no overlap between these two sets.…”

The Complex Action Recognition via the Correlated Topic Model

“…In this paper, we use optical flow to detect the relative direction and magnitude of environmental motion observed in reference to an observer and also describe the movement of object from current image with respect to the last image. The optical flow [22, 29] equation can be assumed to hold for all pixels within a window centered at p , the local image flow (velocity) vector ( V x , V y ) must be satisfied, and we define some equations as follows. $\begin{array}{c}{I}_x\left(s_1\right)V_x+I_y\left(s_1\right)V_y=-I_t\left(s_1\right),\\ I_x\left(s_2\right)V_x+I_y\left(s_2\right)V_y=-I_t\left(s_2\right),\\ ⋮\\ I_x\left(s_d\right)V_x+I_y\left(s_d\right)V_y=-I_t\left(s_d\right),\end{array}$ where s 1 , s 2 ,…, s d are the pixels inside the window and I x ( s i ), I y ( s i ), and I t ( s i ) are the partial derivatives of the image I with respect to position x , y and time t , evaluated at the point S i and the current time.…”

“…In order to capture the correlation of topics, we model the hyperparameter of topic prior distribution as multivariate normal distribution instead of Dirichlet, similar to [29, 32] and the structure topics of dependencies by covariance matrix, and then we use the logistic normal function: $\begin{array}{c}f\left({\theta }_{i}e\right)=\left(\frac{\exp {\theta }_i}{{{\displaystyle false\sum }}_{j=\mathrm{normal1}}^K\exp {\theta }_j}\right)\end{array}$ to project the multivariate normal to topic proportions for each image; here K is the topic number.…”