Abstract. Tele-Immersive applications possess an unusually broad range of networking requirements. As high-speed and Quality of Service-enabled networks emerge, it will becoming more difficult for developers of Tele-Immersion applications, and networked applications in general, to take advantage of these enhanced services. This paper proposes an adaptive networking framework to ultimately allow applications to optimize their network utilization in pace with advances in networking services. In working toward this goal, this paper will present a number of networking techniques for improving performance in tele-immersive applications and examines whether the Differentiated Services mechanism for network Quality of Service is suitable for Tele-Immersion. IntroductionTele-Immersion is the integration of collaborative virtual reality (VR) with audio and video conferencing in the context of data-mining and significant computation. The ultimate goal of Tele-Immersion is not only to reproduce a real face-to-face meeting in every detail, but to provide the "next generation" interface for collaborators, world-wide, to work together in a virtual environment that is seamlessly enhanced by computation and large databases. When participants are Tele-Immersed, they are able to see and interact with each other in a shared virtual environment. They are able to query and visualize data stores and steer complex scientific and engineering simulations [1].One of the challenges of Tele-Immersion is that it poses diverse requirements of the underlying networks ( Figure 1). For example, to convey audio and gestures of virtual participants (avatars,) low network latency is required; to distribute state updates, low latency but reliable data transmission is preferred; and to distribute data sets high-speed bulk data transfer is needed. In this paper, we will present our most recent work in exploiting advanced networking techniques to optimize data distribution in Tele-Immersion. We will describe our experiences in using Quality-ofService-enabled high-speed networks for supporting Tele-Immersion. We will structure this work by proposing an adaptive networking framework to allow application developers to map their data distribution requirements to suitable networking services. We believe that as networking technology becomes more complex, application developers will have to rely increasingly on intelligent adaptive systems to make decisions on how to optimally distribute their data over them. The work discussed in this paper serves as a starting point toward that ultimate goal. Adaptive Networking for Tele-ImmersionWe propose an intelligent adaptive networking system ( Figure 2) consisting of a Strategy Selector, Adaptive Controller and three supporting services: a Resource Monitor, a Quality of Service (QoS) Provisioner and a collection of network transport mechanisms. The Strategy Selector's role is to take application-specified data delivery requirements (e.g. bandwidth, latency, jitter, reliability, etc) and translate them into networki...

ObjectiveA study of supervised automated classification of the cervical vertebrae maturation (CVM) stages using deep learning (DL) network is presented. A parallel structured deep convolutional neural network (CNN) with a pre‐processing layer that takes X‐ray images and the age as the input is proposed.MethodsA total of 1018 cephalometric radiographs were labelled and classified according to the CVM stages. The images were separated according to gender for better model‐fitting. The images were cropped to extract the cervical vertebrae automatically using an object detector. The resulting images and the age inputs were used to train the proposed DL model: AggregateNet with a set of tunable directional edge enhancers. After the features of the images were extracted, the age input was concatenated to the output feature vector. To have the parallel network not overfit, data augmentation was used. The performance of our CNN model was compared with other DL models, ResNet20, Xception, MobileNetV2 and custom‐designed CNN model with the directional filters.ResultsThe proposed innovative model that uses a parallel structured network preceded with a pre‐processing layer of edge enhancement filters achieved a validation accuracy of 82.35% in CVM stage classification on female subjects, 75.0% in CVM stage classification on male subjects, exceeding the accuracy achieved with the other DL models investigated. The effectiveness of the directional filters is reflected in the improved performance attained in the results. If AggregateNet is used without directional filters, the test accuracy decreases to 80.0% on female subjects and to 74.03% on male subjects.ConclusionAggregateNet together with the tunable directional edge filters is observed to produce higher accuracy than the other models that we investigated in the fully automated determination of the CVM stages.

No abstract

No abstract