Momental directional patterns for dynamic texture recognition

Abstract

Understanding the chaotic motions of dynamic textures (DTs) is a challenging problem of video representation for different tasks in computer vision. This paper presents a new approach for an efficient DT representation by addressing the following novel concepts. First, a model of moment volumes is introduced as an effective pre-processing technique for enriching the robust and discriminative information of dynamic voxels with low computational cost. Second, two important extensions of Local Derivative Pattern operator are proposed to improve its performance in capturing directional features. Third, we present a new framework, called Momental Directional Patterns, taking into account the advantages of filtering and local-feature-based approaches to form effective DT descriptors. Furthermore, motivated 1 by convolutional neural networks, the proposed framework is boosted by utilizing more global features extracted from max-pooling videos to improve the discrimination power of the descriptors. Our proposal is verified on benchmark datasets, i.e., UCLA, DynTex, and DynTex++, for DT classification issue. The experimental results substantiate the interest of our method.

Learning graph representation with Randomized Neural Network for dynamic texture classification

“…On the other hand, in [27] the authors introduced the LTGH descriptor, which combines LBPs and gray-level co-occurrence matrix (GLCM) on orthogonal planes (TOP). Nguyen et al [28] proposed the momental directional patterns framework that extends the Local Derivative Pattern operator to improve the capture of directional features. In 4 [29], it is proposed the local tetra pattern operator on three orthogonal planes, which computes feature codes based on the central pixel and directions of the neighbors.…”

Efficient Dynamic Texture Classification with Probabilistic Motifs

“…A slightly modified version of UCLA dataset is often utilized for dynamic texture classification where the original samples into sub-sequences of size 48 × 48. Three popular challenges of this dataset [9], [28], [29] are often used for classification. In the 50-class (4-fold) configuration, a quarter of the data in each class is addressed as testing set and the remaining for the learning.…”

Closed-form camera pose and plane parameters estimation for moments-based visual servoing of planar objects

“…As a solution to these issues, image moments were introduced for visual servoing. [11][12][13][14] Image moments have a broad spectrum of applications in image analysis, such as invariant pattern recognition, 15,16 pose estimation, 17,18 and reconstruction. 19 A set of moments computed from a digital image represents the global characteristics of the image shape and provides much information regarding the different geometrical features of the image.…”