No abstract

Information diffusion prediction is the study of the path of dissemination of news, information, or topics in a structured data such as a graph. Research in this area is focused on two goals, tracing the information diffusion path and finding the members that determine future the next path. The major problem of traditional approaches in this area is the use of simple probabilistic methods rather than intelligent methods. Recent years have seen growing interest in the use of machine learning algorithms in this field. Recently, deep learning, which is a branch of machine learning, has been increasingly used in the field of information diffusion prediction. This paper presents a machine learning method based on the graph neural network algorithm, which involves the selection of inactive vertices for activation based on the neighboring vertices that are active in a given scientific topic. Basically, in this method, information diffusion paths are predicted through the activation of inactive vertices by active vertices. The method is tested on three scientific bibliography datasets: The Digital Bibliography and Library Project (DBLP), Pubmed, and Cora. The method attempts to answer the question that who will be the publisher of the next article in a specific field of science. The comparison of the proposed method with other methods shows 10% and 5% improved precision in DBLP and Pubmed datasets, respectively.

Human action recognition with color and depth sensors has received increasing attention in image processing and computer vision. This paper target is to develop a novel deep model for recognizing human action from the fusion of RGB-D videos based on a Convolutional Neural Network. This work is proposed a novel 3D Convolutional Neural Network architecture that implicitly captures motion information between adjacent frames, which are represented in two main steps: As a First, the optical flow is used to extract motion information from spatio-temporal domains of the different RGB-D video actions. This information is used to compute the features vector values from deep 3D CNN model. Secondly, train and evaluate a 3D CNN from three channels of the input video sequences (i.e. RGB, depth and combining information from both channels (RGB-D)) to obtain a feature representation for a 3D CNN model. For evaluating the accuracy results, a Convolutional Neural Network based on different data channels are trained and additionally the possibilities of feature extraction from 3D Convolutional Neural Network and the features are examined by support vector machine to improve and recognize human actions. From this methods, we demonstrate that the test results from RGB-D channels better than the results from each channel trained separately by baseline Convolutional Neural Network and outperform the state of the art on the same public datasets.

Abstract-In this paper, we present a novel system to analyze human body motions for action recognition task from two sets of features using RGBD videos. The Bag-of-Features approach is used for recognizing human action by extracting local spatialtemporal features and shape invariant features from all video frames. These feature vectors are computed in four steps: Firstly, detecting all interest keypoints from RGB video frames using Speed-Up Robust Features and filters motion points using Motion History Image and Optical Flow, then aligned these motion points to the depth frame sequences. Secondly, using a Histogram of orientation gradient descriptor for computing the features vector around these points from both RGB and depth channels, then combined these feature values in one RGBD feature vector. Thirdly, computing Hu-Moment shape features from RGBD frames; fourthly, combining the HOG features with Hu-moments features in one feature vector for each video action. Finally, the k-means clustering and the multi-class K-Nearest Neighbor is used for the classification task. This system is invariant to scale, rotation, translation, and illumination. All tested, are utilized on a dataset that is available to the public and used often in the community. By using this new feature combination method improves performance on actions with low movement and reach recognition rates superior to other publications of the dataset.

Human action recognition from the videos is one of the most attractive topics in computer vision during the last decades due to wide applications development. This research has mainly focused on learning and recognizing actions from RGB and Depth videos (RGBD). RGBD is a powerful source of data providing the aligned depth information which has great ability to improve the performance of different problems in image understanding and video processing. In this work, a novel system for human action recognition is proposed to extract distinctive spatio and temporal feature vectors for presenting the spatio-temporal evolutions from a set of training and testing video sequences of different actions. The feature vectors are computed in two steps: The First step is the motion detection from all video frames by using spatio-temporal retina model. This model gives a good structuring of video data by removing the noise and illumination variation and is used to detect potentially salient areas, these areas represent the motion information of the moving object in each frame of video sequences. In the Second step, because of human motion can be seen as a type of texture pattern, the local binary pattern descriptor (LBP) is used to extract features from the spatio-temporal salient areas and formulated them as a histogram to make the bag of feature vectors. To evaluate the performance of the proposed method, the k-means clustering, and Random Forest classification is applied on the bag of feature vectors. This approach is demonstrated that our system achieves superior performance in comparison with the state-of-the-art and all experimental results are depending on two public RGBD datasets.