Multimedia Segments Research Articles

Design of a 5G Multimedia Broadcast Application Function Supporting Adaptive Error Recovery

The demand for mobile multimedia streaming services has been steadily growing in recent years. Mobile multimedia broadcasting addresses the shortage of radio resources but introduces a network error recovery problem. Retransmitting multimedia segments that are not correctly broadcast can cause service disruptions and increased service latency, affecting the quality of experience perceived by end users. With the advent of networking paradigms based on virtualization technologies, mobile networks have been enabled with more flexibility and agility to deploy innovative services that improve the utilization of available network resources. This paper discusses how mobile multimedia broadcast services can be designed to prevent service degradation by using the computing capabilities provided by multiaccess edge computing (MEC) platforms in the context of a 5G network architecture. An experimental platform has been developed to evaluate the feasibility of a MEC application to provide adaptive error recovery for multimedia broadcast services. The results of the experiments carried out show that the proposal provides a flexible mechanism that can be deployed at the network edge to lower the impact of transmission errors on latency and service disruptions.

Violence Recognition Based on Auditory-Visual Fusion of Autoencoder Mapping

In the process of violence recognition, accuracy is reduced due to problems related to time axis misalignment and the semantic deviation of multimedia visual auditory information. Therefore, this paper proposes a method for auditory-visual information fusion based on autoencoder mapping. First, a feature extraction model based on the CNN-LSTM framework is established, and multimedia segments are used as whole input to solve the problem of time axis misalignment of visual and auditory information. Then, a shared semantic subspace is constructed based on an autoencoder mapping model and is optimized by semantic correspondence, which solves the problem of audiovisual semantic deviation and realizes the fusion of visual and auditory information on segment level features. Finally, the whole network is used to identify violence. The experimental results show that the method can make good use of the complementarity between modes. Compared with single-mode information, the multimodal method can achieve better results.

Multimedia presentation organization and playout management using intelligent agents

This paper presents a model for organization, computation and management of automated multimedia presentations based on active multimedia segments retrieved from a multimedia information system as well as the Web. Existing multimedia presentation system is extended in such a way that its content selection component spans from local data resources to the entire Web by use of intelligent agents which associates related multimedia segments. We describe a multimedia presentation authoring environment in which a new heuristic method is introduced based on the multimedia resources selected and organization operators to apply. This method is used for the construction of a presentation graph representing an organized presentation. Once the presentation graph is constructed, we show how to obtain an event-point representation of the presentation graph. Based on event-point representation, three methods are given for playing out the constructed presentations at the presentation terminal. Presentation environments of the end-users are modeled as (i) without any constraint, (ii) with a single constraint, and (iii) with multiple constraints (one constraint for each type of multimedia segments in the organized presentation). In accordance with these limitations (i.e., without violating any of the end-user specified constraints), three methods play out any organized presentation.

System and method for indexing videos based on speaker distinction

A system and method for indexing multimedia files utilizes audio characteristics of predefined audio content contained in selected multimedia segments of the multimedia files to distinguish the selected multimedia segments. In the exemplary embodiment, the predefined audio content is speech contained in video segments of video files. Furthermore, the audio characteristics are speaker characteristics. The speech-containing video segments are detected (702) by analyzing the audio contents of the video files. The audio contents of the speech-containing video segments are then characterized (604, 606) to distinguish the video segments according to speakers. The indexing of speech-containing video segments based on speakers allows users to selectively access video segments that contain speech from a particular speaker without having to manually search all the speech-containing video segments.

E-learning and Fluency Disorders

E-learning and Fluency Disorders

A Tale of Two Learning Objects

Learning Objects are digital or non-digital entities, which can be used, re-used or referenced during technology supported learning. These Learning Objects include multimedia segments, interactive learning systems, instructional software tools, course contents, and course management systems. The challenge for education is to design these learning objects. These objects will draw both from knowledge about human cognition and from practical applications of how technology can facilitate complex learning and teaching tasks. This article describes the design and the implementation of two learning objects: the Web Enabled Basic Knowledge Accessibility Program (WEBKAP) and the Web Enabled Automated Test Generation and Management System (WEATGAMS). These tools are created with a common thread: to ameliorate the learning experiences of students.

Multimedia search apparatus and method for searching multimedia content using speaker detection by audio data

A multimedia search apparatus and method for searching multimedia content using speaker detection to segment the multimedia content. The multimedia search apparatus receives a search request from a user device. The search request identifies the target speaker for which the search is to be conducted. Based on the search request, the multimedia search apparatus retrieves multimedia content from a multimedia database. The multimedia search apparatus retrieves models, such as Gaussian Mixture Models (GMMs), from a model storage device, corresponding to the target speaker and background data. Based on the retrieved models, the multimedia search device searches the multimedia data of the multimedia content and segments the multimedia data. The segments are identified by calculating an average normalized score for a block of frames of the multimedia data and determining if the average normalized score for the block of frames exceeds one or more predetermined thresholds.

Constraint-based automation of multimedia presentation assembly

In this paper, we present a methodology for automated construction of multimedia presentations. Semantic coherency of a multimedia presentation is expressed in terms of presentation inclusion and exclusion constraints. When a user specifies a set of segments for a presentation, the multimedia database system adds segments into and/or deletes segments from the set in order to satisfy the inclusion and exclusion constraints. We discuss the consistency and the satisfiability of inclusion and exclusion constraints when exclusion is allowed. Users express a presentation query by (a) pointing and clicking to an initial set of desired multimedia segments to be included into the presentation, and (b) specifying an upper bound on the time length of the presentation. The multimedia database system then finds the set of segments satisfying the inclusion-exclusion constraints and the time bound. Using priorities for segments and inclusion constraints, we give two algorithms for automated presentation assembly and discuss their complexity. To automate the assembly of a presentation with concurrent presentation streams, we introduce presentation organization constraints that are incorporated into the multimedia data model, independent of any presentation. We define four types of presentation organization constraints that, together with an underlying database ordering, allow us to obtain a unique presentation graph for a given set of multimedia segments. We briefly summarize a prototype system that fully incorporates the algorithms for the segment selection problem.

User Control and Dynamic Reorganization of Multimedia Presentations During Playout

A methodology to dynamically adjust the multimedia presentations as a result of user interactions during the playout is introduced. A constraint-driven approach is described for the automated assembly, organization and playout of presentations from multimedia databases. To allow interactive user control over the playout of an automatically generated presentation, a set of control buttons are provided. Using these buttons, the user can control and change the flow of the presentation in a variety of ways. Interactive user control during the presentation playout, however, can introduce new delays in the original presentation schedule of multimedia segments and possibly change playout constraints. To account for these delays and changes, strategies to (a) dynamically reorganize the subpresentations into a presentation, (b) replace segments with their compacted clips, and (c) delete segments on the presentation graph are proposed.

A Constraint-Driven Approach to Automate the Organization andPlayout of Presentations in Multimedia Databases

We introduce a constraint-driven methodology for the automated assembly, organization and playout of presentations from multimedia databases. We use inclusion and exclusion constraints for extracting a semantically coherent set of multimedia segments. Presentation organization constraints are utilized for organizing the multimedia segments into a presentation, which in turn helps decide the playout order of the extracted multimedia segments. The playout order of the segments is represented in a presentation graph. If the specified set of organization constraints are not sufficient to construct a unique presentation graph, we propose two techniques so that a unique graph is constructible. We also propose two playout algorithms, one for the generation, start and termination of playout agents, the other for dynamic control of playout management on organized presentations. The characteristics of these algorithms are expressed in terms of presentation playout parameters.

Robert Desrosiers: Breaking Boundaries, Shaking Foundations

The work of the Desrosiers Dance Theatre is a mind-bending, breath-taking combination of dance, theatre, original music, mime, tai-chi, acrobatics, multi-media segments, elaborate sets, costumes and props and startling special effects. Based in Toronto, and housed in a new studio-office space in the East End of the city, the company has toured nationally and internationally during the first decade of its existence and now, as it prepares to enter the second daring decade, a New York City premiere (at the prestigious City Center) and a new full-length work this fall in Toronto loom large.