Video Content Classification Research Articles

A Short Video Classification Framework Based on Cross-Modal Fusion.

The explosive growth of online short videos has brought great challenges to the efficient management of video content classification, retrieval, and recommendation. Video features for video management can be extracted from video image frames by various algorithms, and they have been proven to be effective in the video classification of sensor systems. However, frame-by-frame processing of video image frames not only requires huge computing power, but also classification algorithms based on a single modality of video features cannot meet the accuracy requirements in specific scenarios. In response to these concerns, we introduce a short video categorization architecture centered around cross-modal fusion in visual sensor systems which jointly utilizes video features and text features to classify short videos, avoiding processing a large number of image frames during classification. Firstly, the image space is extended to three-dimensional space-time by a self-attention mechanism, and a series of patches are extracted from a single image frame. Each patch is linearly mapped into the embedding layer of the Timesformer network and augmented with positional information to extract video features. Second, the text features of subtitles are extracted through the bidirectional encoder representation from the Transformers (BERT) pre-training model. Finally, cross-modal fusion is performed based on the extracted video and text features, resulting in improved accuracy for short video classification tasks. The outcomes of our experiments showcase a substantial superiority of our introduced classification framework compared to alternative baseline video classification methodologies. This framework can be applied in sensor systems for potential video classification.

Chaser Priori Wolf (CPW) Optimization an Improved Optimization Technique Video Content Classification and Detection

Optimizers play a crucial role in video object detection by promoting the training and improving the performance of the model. Optimizers are responsible for minimizing the loss function during training. The parameters of models are updated iteratively based on the gradients of the loss parameters. By continuously adjusting the parameters in the direction of the steepest descent, optimizers guide the model towards convergence, reducing the loss and improving the object detection performance. In the proposed paper hybrid optimizer named chaser priori wolf optimizer is proposed. The chaser priori wolf optimization is based on the hybridization of cat swarm optimization and coyote optimization. Well-known optimizers like SGD, ADAM, adagrad, adadelta and RMSprop are used as default optimizers by researchers. The proposed work introduced CPW optimizer which works for classification to improve the convergence and feature selection. The comparative result showed an increase in the performance of CNN based YOLO model. The results are compared concerning sensitivity, specificity and accuracy. Results clearly showed improvement in all performance metrics and the average improvement in comparison with state of art architecture is 10.3%.

Evaluation of the content of YouTubeTM videos about local anesthesia in pediatric dentistry

Abstract
 Aim: The purpose of this study was to assay the quality and substance of videos on YouTubeTM about local anesthesia in children and to assay whether they are practical for patients and parents.
 Materials and Methods: A research was applied on YoutubeTM using the search term "local anesthesia in pediatric dentistry" with the assumed sorting set to "sort by relevance". In our study, 113 videos out of 213 were excluded and 100 videos were analyzed. For video content classification, a 23-point score scale was used to classify video groups as low, moderate, and high video context.
 Results: Most of the videos were uploaded by healthcare professionals (n=93; 93%), while the rest were uploaded by healthcare companies/websites (n=1; 1%) and others (tv channels, news agencies) (n=6; 6%). The average duration of videos was 9.49 minutes (range: 0.5-57.57 minutes; median: 4.25). 
 Conclusion: Although there are many videos on local anesthesia in children on YouTubeTM, these videos contain limited information in terms of content. There is a requirement to develop the quality of YouTubeTM videos on this subject by dentists.

ENCVIDC: an innovative approach for encoded video content classification

ENCVIDC: an innovative approach for encoded video content classification

An Automatic Classification Method of Sports Teaching Video Using Support Vector Machine

There are many different types of sports training films, and categorizing them can be difficult. As a result, this research introduces an autonomous video content classification system that makes managing large amounts of video data easier. This research provides a video feature extraction approach using a support vector machine (SVM) video classification algorithm and a mix of video and audio dual-mode characteristics. It automates the classification of cartoons, ads, music, news, and sports videos, as well as the detection of terrorist and violent moments in films. To begin, a new feature expression scheme, the MPEG-7 visual descriptor subcombination, is proposed based on an analysis of the existing video classification algorithms, with the goal of addressing the problems in these algorithms. This is accomplished by analyzing the visual differences of the five video classification algorithms. The model was able to extract 9 descriptors from the four characteristics of color, texture, shape, and motion, resulting in a new overall visual feature with good results. The results suggest that the algorithm optimizes video segmentation by highlighting disparities in feature selection between different categories of films. Second, the support vector machine’s multivideo classification performance is improved by the enhanced secondary prediction method. Finally, a comparison experiment with current related similar algorithms was conducted. The suggested method outperformed the competition in the accuracy of video classification in five different types of videos, as well as in the recognition of terrorist and violent incidents.

Research on Image-Based Movement Accuracy Monitoring of Aerobics

The movement accuracy monitoring of aerobics is mostly performed through three-dimensional reconstruction of aerobic movements. The feature extraction of aerobics is based on the optimal classification decision function, which extracts all the features of aerobics and thus reduces the accuracy of aerobics monitoring. In order to extract the aerobic motion in the background with higher accuracy, a new image-based monitoring method is proposed. First, the Kinect depth image acquisition method is used to preprocess the image, and then Hog3D is used to extract aerobic movement features and analyze the extraction results. This new method solves the problem of video content classification in aerobics precision monitoring. The Adaboost method in probability statistics is used to identify the accuracy of aerobic motions. This paper uses probability function to link the postures of aerobics and forms an action sequence and its ergodic function to take the maximum value of an aerobic exercise. The accuracy of aerobics is monitored by using the method of level by level proportional example. The experimental results show that this method can effectively improve the accuracy of aerobic track monitoring, reduce the energy consumption of aerobic movement accuracy monitoring, and has good use value.

Video mining for facial action unit classification using statistical spatial–temporal feature image and LoG deep convolutional neural network

In this paper, we aim to improve the convolutional deep learning by means of proposed statistical feature image descriptor and Laplacian of Gaussian filtering. We propose a statistical feature image descriptor (SFID) that is composed of spatial and temporal parts for video content classification using convolutional deep neural network. We apply the proposed descriptor to multi-view action unit classification. The SFID is a statistical representation of the raw image based upon K-abstraction levels. It is capable of addressing the fixing input array size of the deep learning model. Further, it eliminates redundancy in representation of the content; hence, it reduces computation cost. The proposed SFID can be in spatial and/or temporal form. The temporal form is particularly important in video content classification. We added a new layer of Laplacian of Gaussian filter (LoG) right before fully connected layer into the regular deep convolutional neural network (DCNN) structure. The parameters of the LoG are adaptively calculated using the Gaussian mixture models. The classification results are compared with regular DCNN, SVM models, and KNN together with feature descriptors of SIFT and SURF. The results show that the proposed feature descriptor and introducing a LoG filter layer give promising performance for deep learning.

Approaches to Detect Online Radicalization: A Survey

Approaches to Detect Online Radicalization: A Survey

Text-based video content classification for online video-sharing sites

With the emergence of Web 2.0, sharing personal content, communicating ideas, and interacting with other online users in Web 2.0 communities have become daily routines for online users. User-genera...

An automatic video content classification scheme based on combined visual features model with modified DAGSVM

Automatic video content classification attracts much attention from researchers in multimedia analysis because the management of video content is a challenging task. In this paper, a visual feature representation composed of editing, color, texture and motion features is proposed which is shown to be effective in differentiating among various video contents. A modified Directed Acyclic Graph Support Vector Machine (DAGSVM) model as the classifier is also presented. Experiments show that the features extracted have improved the discriminative ability between different video contents and the computational complexity has also been reduced. By introducing the DAG policy, the performance of the classifier has been enhanced and the classification results demonstrate the precision and effectiveness of this approach, compared with the other two classification methods. In addition, the proposed algorithm can be applied to video searching and harmful-video content filtering, etc.

Text‐based video content classification for online video‐sharing sites

AbstractWith the emergence of Web 2.0, sharing personal content, communicating ideas, and interacting with other online users in Web 2.0 communities have become daily routines for online users. User‐generated data from Web 2.0 sites provide rich personal information (e.g., personal preferences and interests) and can be utilized to obtain insight about cyber communities and their social networks. Many studies have focused on leveraging user‐generated information to analyze blogs and forums, but few studies have applied this approach to video‐sharing Web sites. In this study, we propose a text‐based framework for video content classification of online‐video sharing Web sites. Different types of user‐generated data (e.g., titles, descriptions, and comments) were used as proxies for online videos, and three types of text features (lexical, syntactic, and content‐specific features) were extracted. Three feature‐based classification techniques (C4.5, Naïve Bayes, and Support Vector Machine) were used to classify videos. To evaluate the proposed framework, user‐generated data from candidate videos, which were identified by searching user‐given keywords on YouTube, were first collected. Then, a subset of the collected data was randomly selected and manually tagged by users as our experiment data. The experimental results showed that the proposed approach was able to classify online videos based on users' interests with accuracy rates up to 87.2%, and all three types of text features contributed to discriminating videos. Support Vector Machine outperformed C4.5 and Naïve Bayes techniques in our experiments. In addition, our case study further demonstrated that accurate video‐classification results are very useful for identifying implicit cyber communities on video‐sharing Web sites.

Video content classification based on 3-D Eigen analysis

To achieve video understanding, it is of utmost practical importance to classify videos according to its spatial and temporal features in an efficient and effective manner. It still remains, however, largely an elusive task. In still-image analysis, thanks to the great efforts made by many researchers, a broad spectrum of methods have been developed with great success, especially the ones based on eigen analysis due to its efficacy. In this paper, inspired by the impressive performance achieved by this framework, we will develop a content-based video classification method based on three-dimensional (3-D) eigen analysis. Unlike most other video understanding schemes where the spatial and temporal contents play different roles in the processing, this new method treats a video as a solid within a 3-D Euclidean space and can, thus, naturally take advantage of the spatial and temporal contents existing in videos. After computing the eigen values and corresponding eigen vectors of the autocorrelation matrix for each small 3-D macroblock, different labels are assigned regarding its spatial/temporal natures based on the behavioral properties of the eigen values and eigen vectors. Extensive empirical studies have suggested encouraging performance for the use of this eigen analysis-based video classification method.

Context and memory in multimedia content analysis

With the advent of broadband networking, video will be available online as well as through traditional distribution channels. The merging of entertainment and information media makes video content classification and retrieval a necessary tool. To provide fast retrieval, content management systems must discern between categories of video. Automatic multimedia analysis techniques for deriving high-level descriptions and annotations have experienced a tremendous surge in interest. Academia and industry have also been challenged to develop realistic applications-from home media library organizers and multimedia lecture archives to broadcast TV content navigators and video-on-demand-in pursuit of the killer application. Current content classification technologies have undoubtedly emerged from traditional image processing and computer vision, audio analysis and processing, and information retrieval. Although terminology varies, the algorithms generally fall into three categories: tangible detectors, high-level abstractors, and latent or intangible descriptors. This paper presents the reflections of the work done by the author and the work ahead.

MPEG-7 based description schemes for multi-level video content classification

MPEG-7 has emerged as the standard for multimedia data content description for efficiently describing multimedia content. In this context, its primary goal is to provide flexible and effective searching and retrieval of multimedia resources. Most of the earlier work on MPEG-7 description schemes (DSs) and descriptors (Ds) focuses on the description of a single multimedia document, whereas MPEG-7 can be further exploited to support more advances implementations under multimedia database systems. Therefore, it is important to reconsider issues related to high level multimedia modeling and representation, in the light of the MPEG-7 perspective. In this paper, we propose a high level multimedia representation and description scheme based on multi-level video modeling and semantic video classification. The proposed multi-level multimedia representation and DSs are expected to support more effective video content indexing and accessing operations. The presented DSs and Ds are further described by using the XML Schema language, which has been adopted as the basis of the Description Definition Language (DDL) of the MPEG-7 standard.

VIDEO CLASSIFICATION USING OBJECT TRACKING

Content description becomes important in the ubiquity of video content on the Web and consumer devices. Video classification is needed so that more appropriate description and search methods can be applied. This paper describes two methods for video content classification: a Nearest Neighbor (NN) method relying on domain knowledge and Hidden Markov Model (HMM) based method. Our approach stems from the observation that in different TV categories, there are different objects (e.g., face and text) trajectory patterns. Face and text tracking is applied to video segments to extract face and text trajectories. We used NN and HMM to classify a given video segment into predefined classes, e.g., commercial, news, situation comedy and soap. Our preliminary experimental results show classification accuracy of 75% for NN and over 80% for HMM based method on short video segments.

Motion recovery for video content classification

Like other types of digital information, video sequences must be classified based on the semantics of their contents. A more-precise and completer extraction of semantic information will result in a more-effective classification. The most-discernible difference between still images and moving pictures stems from movements and variations. Thus, to go from the realm of still-image repositories to video databases, we must be able to deal with motion. Particularly, we need the ability to classify objects appearing in a video sequence based on their characteristics and features such as shape or color, as well as their movements. By describing the movements that we derive from the process of motion analysis, we introduce a dual hierarchy consisting of spatial and temporal parts for video sequence representation. This gives us the flexibility to examine arbitrary sequences of frames at various levels of abstraction and to retrieve the associated temporal information (say, object trajectories) in addition to the spatial representation. Our algorithm for motion detection uses the motion compensation component of the MPEG video-encoding scheme and then computes trajectories for objects of interest. The specification of a language for retrieval of video based on the spatial as well as motion characteristics is presented.