Nokia (Canada)

We present an efficient algorithm for continuous image recognition and feature descriptor tracking in video which operates by reducing the search space of possible interest points inside of the scale space image pyramid. Instead of performing tracking in 2D images, we search and match candidate features in local neighborhoods inside the 3D image pyramid without computing their feature descriptors. The candidates are further validated by fitting to a motion model. The resulting tracked interest points are more repeatable and resilient to noise, and descriptor computation becomes much more efficient because only those areas of the image pyramid that contain features are searched. We demonstrate our method on real-time object recognition and label augmentation running on a mobile device.

Although there has been much interest in computational photography within the research and photography communities, progress has been hampered by the lack of a portable, programmable camera with sufficient image quality and computing power. To address this problem, we have designed and implemented an open architecture and API for such cameras: the Frankencamera. It consists of a base hardware specification, a software stack based on Linux, and an API for C++. Our architecture permits control and synchronization of the sensor and image processing pipeline at the microsecond time scale, as well as the ability to incorporate and synchronize external hardware like lenses and flashes. This paper specifies our architecture and API, and it describes two reference implementations we have built. Using these implementations we demonstrate six computational photography applications: HDR viewfinding and capture, low-light viewfinding and capture, automated acquisition of extended dynamic range panoramas, foveal imaging, IMU-based hand shake detection, and rephotography. Our goal is to standardize the architecture and distribute Frankencameras to researchers and students, as a step towards creating a community of photographer-programmers who develop algorithms, applications, and hardware for computational cameras.

In this paper, we identify mobile social networks as an important new direction of research in mobile computing, and show how an expanded definition of mobile social networks that includes sensor networks can enable exciting new context-aware applications, such as context-aware video screens, music jukeboxes, and mobile health applications. We offer SocialFusion as a system capable of systematically integrating such diverse mobile, social, and sensing input streams and effectuating the appropriate context-aware output action. We explain some of the major challenges that SocialFusion must overcome. We describe some preliminary results that we have obtained in implementing the SocialFusion vision.

Recent experimental studies have shown that wireless links in real sensor networks can be extremely unreliable, deviating to a large extent from the idealized perfect-reception-within-range models used in common network simulation tools. Previously proposed geographic routing protocols commonly employ a maximum-distance greedy forwarding technique that works well in ideal conditions. However, such a forwarding technique performs poorly in realistic conditions as it tends to forward packets on lossy links. Based on a recently developed link loss model, we study the performance of a wide array of forwarding strategies, via analysis, extensive simulations and a set of experiments on motes. We find that the product of the packet reception rate and the distance improvement towards destination ( PRR × d ) is a highly suitable metric for geographic forwarding in realistic environments.

Opportunistic routing is a new routing paradigm that takes advantage of the broadcast characteristic of a wireless channel for data delivery in a wireless mesh network. Network coding has recently emerged as a new coding paradigm that can significantly improve the throughput performance of a WMN. In this article we explore the combination of opportunistic routing and network coding for improving the performance of a WMN. We first review existing opportunistic routing and coding-aware routing protocols, respectively, classify these protocols based on different criteria, and discuss their merits and drawbacks. We then propose a coding-aware opportunistic routing mechanism that combines hop-by-hop opportunistic forwarding and localized inter-flow network coding for improving the throughput performance of a WMN. Through opportunistic forwarding, CORE allows the next-hop node with the most coding gain to continue the packet forwarding. Through localized network coding, CORE attempts to maximize the number of packets that can be carried in a single transmission. Simulation results show that CORE can significantly improve the throughput performance of a WMN as compared with existing protocols.

A nonphotorealistic algorithm for retargeting images adapts large images so that important objects in the image are still recognizable when displayed at a lower target resolution. Unlike existing image manipulation techniques such as cropping and scaling, the retargeting algorithm can handle multiple important objects in an image. To identify the important objects in an image, we must first segment the image. We use mean-shift image segmentation to decompose an image into homogeneous regions.

The theory of propagating waves near a surface is reviewed with an eye to gain insight into the mechanisms involved, and to provide analytical-based models, for power-efficient on-body propagation. The Zenneck wave, and in particular the Norton wave, are appraised as candidate mechanisms for the propagation. For flush-mounted (ldquoband aidrdquo) antennas, desired for on-body sensors, the Norton wave is the only direct propagation mechanism between the sensors. The Norton wave fits very well to simulation results presented here, and comparisons are also made with available published physical experiments, although these measurements typically feature the optical paths of elevated, or non-flush, antennas.

An essential aspect of learning a second language is the acquisition of vocabulary. However, acquiring vocabulary is often a protracted process that requires repeated and spaced exposure; which can be difficult to accommodate given the busyness of daily living. In this paper, we explore if a learner can implicitly acquire second language vocabulary through her explicit interactions with her mobile phone (e.g., navigating multiple home screens) using an interface we developed called Vocabulary Wallpaper. In addition, we examine if the type of vocabulary this technique exposes to the learner, whether it is contextually relevant or contextually-independent will influence the learner's rate of vocabulary acquisition. The results of our study show participants were able to use Vocabulary Wallpaper to increase the number of second language vocabulary that they can recognize and recall and their rate of vocabulary acquisition was significantly greater when presented with a contextually relevant vocabulary than a contextually-independent vocabulary.

IntroductionWhereas traditional marketing has concentrated on the value and competitive advantages of a product or service, contemporary marketing takes a holistic approach by also considering the purchasing process and the retail store atmosphere to evoke a positive shopping experience (Levy and Weitz, 2009). Neuromarketing has surfaced as a new branch of marketing that

Abstract The viewfinder of a digital camera has traditionally been used for one purpose: to display to the user a preview of what is seen through the camera's lens. High quality cameras are now available on devices such as mobile phones and PDAs, which provide a platform where the camera is a programmable device, enabling applications such as online computational photography, computer vision‐based interactive gaming, and augmented reality. For such online applications, the camera viewfinder provides the user's main interaction with the environment. In this paper, we describe an algorithm for aligning successive viewfinder frames. First, an estimate of inter‐frame translation is computed by aligning integral projections of edges in two images. The estimate is then refined to compute a full 2D similarity transformation by aligning point features. Our algorithm is robust to noise, never requires storing more than one viewfinder frame in memory, and runs at 30 frames per second on standard smartphone hardware. We use viewfinder alignment for panorama capture, low‐light photography, and a camera‐based game controller.

Standardization activities are nearly complete for single wavelength 25 Gb/s time-division multiplexed (TDM) passive optical networks (PONs) and well underway for 50 Gb/s TDM PONs. There is considerable debate in the industry about which technology will be the “next step” after 10 Gb/s TDM PON, now finally starting to ramp up to mass deployment. 50 Gb/s PON clearly brings a <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"> <mml:mrow class="MJX-TeXAtom-ORD"> <mml:mn>2</mml:mn> </mml:mrow> <mml:mo>×</mml:mo> </mml:math> bandwidth advantage over 25 Gb/s, at least in the downstream direction. On the other hand, the increase of speed to 50 Gb/s brings with it a substantial receiver sensitivity penalty of at least 4 dB, which has a chain effect on transceiver architecture, cost, and time-to-market. In this paper, each of those elements is investigated, quantified, and compared to 25 Gb/s.

Although there has been much interest in computational photography within the research and photography communities, progress has been hampered by the lack of a portable, programmable camera with sufficient image quality and computing power. To address this problem, we have designed and implemented an open architecture and API for such cameras: the Frankencamera. It consists of a base hardware specification, a software stack based on Linux, and an API for C++. Our architecture permits control and synchronization of the sensor and image processing pipeline at the microsecond time scale, as well as the ability to incorporate and synchronize external hardware like lenses and flashes. This paper specifies our architecture and API, and it describes two reference implementations we have built. Using these implementations we demonstrate six computational photography applications: HDR viewfinding and capture, low-light viewfinding and capture, automated acquisition of extended dynamic range panoramas, foveal imaging, IMU-based hand shake detection, and rephotography. Our goal is to standardize the architecture and distribute Frankencameras to researchers and students, as a step towards creating a community of photographer-programmers who develop algorithms, applications, and hardware for computational cameras.

In recent years, the CHI community has seen significant growth in research on Human-Centered Responsible Artificial Intelligence. While different research communities may use different terminology to discuss similar topics, all of this work is ultimately aimed at developing AI that benefits humanity while being grounded in human rights and ethics, and reducing the potential harms of AI. In this special interest group, we aim to bring together researchers from academia and industry interested in these topics to map current and future research trends to advance this important area of research by fostering collaboration and sharing ideas.

This paper analyzes the potential of cooperative proxy caching for peer-to-peer (P2P) traffic as a means to ease the burden imposed by P2P traffic on Internet Service Providers (ISPs). In particular, we propose two models for cooperative caching of P2P traffic. The first model enables cooperation among caches that belong to different autonomous systems (ASs), while the second considers cooperation among caches deployed within the same AS. We analyze the potential gain of cooperative caching in these two models. To perform this analysis, we conduct an eight-month measurement study on a popular P2P system to collect traffic traces for multiple caches. Then, we perform extensive trace-based simulations to analyze different angles of cooperative caching schemes. Our results demonstrate that: 1) significant improvement in byte hit rate can be achieved using cooperative caching, 2) simple object replacement policies are sufficient to achieve that gain, and 3) the overhead imposed by cooperative caching is negligible. In addition, we develop an analytic model to assess the gain from cooperative caching in different settings. The model accounts for number of caches, salient P2P traffic features, and network characteristics. Our model confirms that substantial gains from cooperative caching are attainable under wide ranges of traffic and network characteristics.

We assess the status of current generation 25G and 50G time division multiplexed passive optical network (TDM PON) technologies based on leveraging the cost efficiencies of the Ethernet intra-datacenter ecosystem. As a first step towards 100G TDM PON, we predict the real-world impact of a flexible modulation enhancement to 50G PON, whereby four-level pulse amplitude modulation (PAM4) symbols can be transmitted at the same symbol rate as 50 Gb/s PAM2, but only where excess margins permit. We find that sufficient margins are likely to exist to allow for a majority of future 50G PON optical network units to operate at 100 Gb/s PAM4. Next, we look at the options for a 100G PON capable of supporting the full loss budget and reach requirements. There is no technical risk if coherent technology is adopted, but intensity-modulation and direct-detection (IM-DD) will provide lower complexity, lower cost, and lower power dissipation. We evaluate this option and conclude that by following IM-DD Ethernet optics to 100 GBd, single wavelength IM-DD will continue to be feasible for 100G PON and will be a strong contender for the next generation of PON after 50 Gb/s.

Augmented and Virtual Reality promises have yet to be met, it remains inhibited by an average user experience. Limits in local processing (headset/handset), form factors and display/optical limitations have kept the technology from reaching its full potential. 5G and Cloud-based processing offers the potential to improve performance, comfort/ergonomics, mobility and QoE barriers. Our research demonstrates a methodology for assessing AR/VR QoE performance on mobile networks and how 5G and cloud-based processing create a better user experience, accelerating the growth of this nascent market. QoE models are proposed for AR/VR which define the key quality indicators (KQIs) for each AR/VR application and we correlate the KQIs performance to network KPIs. The study demonstrates how typical LTE KPIs are insufficient to deliver an immersive QoE for AR/VR. Finally, we quantify the increase in QoE with 5G and edge cloud based on the expected KPI performance improvements. The proposed AR/VR QOE models are evaluated in the context of the enhanced network performance with 5G and Edge Cloud deployment.

For a wide range of smart agriculture use cases, the prospects of utilizing the Internet of Things (IoT) are immense. Many IoT devices can be deployed for precision farming, soil management, automated irritation, information gathering, or performing some local processing to provide various services. Due to the computational capacity limitation of IoT devices and their limited power, UAV-aided mobile-edge computing (MEC) is proposed to provide IoT nodes with additional resources by hosting their computation functions and making smart agriculture use cases more operational. On the other hand, from the implementation viewpoint, Network Function Virtualization (NFV) is an emerging approach recently proposed for flexible management of such computation functions in UAVs and MEC-server. However, efficient orchestration of the virtualized functions is a challenge. In this paper, we consider a decentralized UAV-aided MEC system in which the NFV-enabled processing nodes manage the computational tasks. To be more specific, we consider the smart agriculture use cases that need live streaming/analysis, such as surveillance or environmental monitoring. In such a network, we propose a method for orchestrating the NFVs efficiently, while the network energy consumption throughout the network is minimized. This problem becomes even more crucial when considering a strict condition on the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">instantaneous</i> AoI values. Therefore, the problem is first formulated as a Decentralized <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Constrained</i> Multi-agent Markov Decision Process (Dec-CMMDP). As the formulated problem is NEXP, in the next step, we exploit some structural features of the considered network and introduce the concept of symmetry to simplify the problem. Then, inspired by Augmented Lagrangian dual optimization, a novel decentralized, federated learning-based solution is proposed to solve the problem. Simulation results show the effectiveness of the proposed approach in minimizing the total network energy consumption, minimizing the average AoI, and satisfying the strict condition of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AoI</i> < 100 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">msec</i> for supporting real-time applications in our network parameter settings.

Network Intrusion Detection Systems (NIDS) have been extensively investigated by monitoring real network traffic and analyzing suspicious activities. However, there are limitations in detecting specific types of attacks with NIDS, such as Advanced Persistent Threats (APT). Additionally, NIDS is restricted in observing complete traffic information due to encrypted traffic or a lack of authority. To address these limitations, a Host-based Intrusion Detection system (HIDS) evaluates resources in the host, including logs, files, and folders, to identify APT attacks that routinely inject malicious files into victimized nodes. In this study, a hybrid network intrusion detection system that combines NIDS and HIDS is proposed to improve intrusion detection performance. The host data undergoes a Language Processing (NLP)-based Bidirectional Encoder Representations from Transformers (BERT) model from textual representation to a numerical one in order to process host data in a similar way to the network flow data through machine learning models. The feature flattening technique is applied to flatten two-dimensional host-based features that is provided by BERT into one-dimensional vectors so that host-based and network flow-based features can be processed by advanced Machine Learning (ML) models. In order to enhance HIDS effectiveness, a two-stage collaborative classifier is utilized, which applies two tiers of machine learning algorithms, binary and multi-class classifiers, to detect network intrusions. Once a binary classifier is used to detect benign samples to reduce the complexity of the original problem, the attack data are classified by a multi-class supervised learner to identify attack types. Hence, the overall performance of the two-stage collaborative model outperforms the baseline classifier, XGBoost. The proposed method is shown to generalize across two well-known datasets, CICIDS 2018 and NDSec-1. The performance of XGBoost, which represents conventional ML, is evaluated. Combining host and network features enhances attack detection performance (macro average F1 score) by 8.1% under the CICIDS 2018 dataset and 3.7% under the NDSec-1 dataset. Meanwhile, the two-stage collaborative classifier improves detection performance for most single classes, especially for DoS-LOIC-UDP and DoS-SlowHTTPTest, with improvements of 30.7% and 84.3%, respectively, when compared with the traditional ML models.

Abstract When creating a High‐Dynamic‐Range (HDR) image from a sequence of differently exposed Low‐Dynamic‐Range (LDR) images, the set of LDR images is usually generated by sampling the space of exposure times with a geometric progression and without explicitly accounting for the distribution of irradiance values of the scene. We argue that this choice can produce sub‐optimal results both in terms of the number of acquired pictures and the quality of the resulting HDR image. This paper presents a method to estimate the full irradiance histogram of a scene, and a strategy to select the set of exposures that need to be acquired. Our selection usually requires a smaller or equal set of LDRs, yet produces higher quality HDR images.

The goal of this paper is to summarize research in designing and developing an application framework for intelligent and mobile agents. Agents are the next significant software abstraction; they will soon be as ubiquitous as graphical user interfaces are today. The major contribution of the work described here is a reusable design and implementation of an architecture that addresses all levels of agent behavior. The design is robust and well founded, based on object oriented design patterns. The framework has been employed in preliminary applications in network management and enterprise integration. However, agent systems are still evolving; to facilitate future development, the framework is documented with patterns.