• Achievable Rates of Secure Transmission in Gaussian MISO Channel with Imperfect Main Channel Estimation

      Zhou, Xinyu; Rezki, Zouheir; Alomair, Basel; Alouini, Mohamed-Slim (IEEE, 2016-02-26)
      A Gaussian multiple-input single-output (MISO) fading channel is considered. We assume that the transmitter, in addition to the statistics of all channel gains, is aware instantaneously of a noisy version of the channel to the legitimate receiver. On the other hand, the legitimate receiver is aware instantaneously of its channel to the transmitter, whereas the eavesdropper instantaneously knows all channel gains. We evaluate an achievable rate using a Gaussian input without indexing an auxiliary random variable. A sufficient condition for beamforming to be optimal is provided. When the number of transmit antennas is large, beamforming also turns out to be optimal. In this case, the maximum achievable rate can be expressed in a simple closed form and scales with the logarithm of the number of transmit antennas. Furthermore, in the case when a noisy estimate of the eavesdropper's channel is also available at the transmitter, we introduce the SNR difference and the SNR ratio criterions and derive the related optimal transmission strategies and the corresponding achievable rates.
    • Robust Manhattan Frame Estimation From a Single RGB-D Image

      Bernard Ghanem; Heilbron, Fabian Caba; Niebles, Juan Carlos; Thabet, Ali Kassem (IEEE, 2015-06-02)
      This paper proposes a new framework for estimating the Manhattan Frame (MF) of an indoor scene from a single RGB-D image. Our technique formulates this problem as the estimation of a rotation matrix that best aligns the normals of the captured scene to a canonical world axes. By introducing sparsity constraints, our method can simultaneously estimate the scene MF, the surfaces in the scene that are best aligned to one of three coordinate axes, and the outlier surfaces that do not align with any of the axes. To test our approach, we contribute a new set of annotations to determine ground truth MFs in each image of the popular NYUv2 dataset. We use this new benchmark to experimentally demonstrate that our method is more accurate, faster, more reliable and more robust than the methods used in the literature. We further motivate our technique by showing how it can be used to address the RGB-D SLAM problem in indoor scenes by incorporating it into and improving the performance of a popular RGB-D SLAM method.
    • On the Relationship between Visual Attributes and Convolutional Networks

      Castillo, Victor; Ghanem, Bernard; Niebles, Juan Carlos (IEEE, 2015-06-02)
      One of the cornerstone principles of deep models is their abstraction capacity, i.e. their ability to learn abstract concepts from ‘simpler’ ones. Through extensive experiments, we characterize the nature of the relationship between abstract concepts (specifically objects in images) learned by popular and high performing convolutional networks (conv-nets) and established mid-level representations used in computer vision (specifically semantic visual attributes). We focus on attributes due to their impact on several applications, such as object description, retrieval and mining, and active (and zero-shot) learning. Among the findings we uncover, we show empirical evidence of the existence of Attribute Centric Nodes (ACNs) within a conv-net, which is trained to recognize objects (not attributes) in images. These special conv-net nodes (1) collectively encode information pertinent to visual attribute representation and discrimination, (2) are unevenly and sparsely distribution across all layers of the conv-net, and (3) play an important role in conv-net based object recognition.
    • Doppler time-of-flight imaging

      Heide, Felix; Wetzstein, Gordon; Hullin, Matthias; Heidrich, Wolfgang (Association for Computing Machinery (ACM), 2015-07-30)
      Over the last few years, depth cameras have become increasingly popular for a range of applications, including human-computer interaction and gaming, augmented reality, machine vision, and medical imaging. Many of the commercially-available devices use the time-of-flight principle, where active illumination is temporally coded and analyzed on the camera to estimate a per-pixel depth map of the scene. In this paper, we propose a fundamentally new imaging modality for all time-of-flight (ToF) cameras: per-pixel velocity measurement. The proposed technique exploits the Doppler effect of objects in motion, which shifts the temporal frequency of the illumination before it reaches the camera. Using carefully coded illumination and modulation frequencies of the ToF camera, object velocities directly map to measured pixel intensities. We show that a slight modification of our imaging system allows for color, depth, and velocity information to be captured simultaneously. Combining the optical flow computed on the RGB frames with the measured metric axial velocity allows us to further estimate the full 3D metric velocity field of the scene. We believe that the proposed technique has applications in many computer graphics and vision problems, for example motion tracking, segmentation, recognition, and motion deblurring.
    • ActivityNet: A Large-Scale Video Benchmark for Human Activity Understanding

      Heilbron, Fabian Caba; Castillo, Victor; Ghanem, Bernard; Niebles, Juan Carlos (IEEE, 2015-06-02)
      In spite of many dataset efforts for human action recognition, current computer vision algorithms are still severely limited in terms of the variability and complexity of the actions that they can recognize. This is in part due to the simplicity of current benchmarks, which mostly focus on simple actions and movements occurring on manually trimmed videos. In this paper we introduce ActivityNet, a new largescale video benchmark for human activity understanding. Our benchmark aims at covering a wide range of complex human activities that are of interest to people in their daily living. In its current version, ActivityNet provides samples from 203 activity classes with an average of 137 untrimmed videos per class and 1.41 activity instances per video, for a total of 849 video hours. We illustrate three scenarios in which ActivityNet can be used to compare algorithms for human activity understanding: untrimmed video classification, trimmed activity classification and activity detection.
    • Fast and Flexible Convolutional Sparse Coding

      Heide, Felix; Heidrich, Wolfgang; Wetzstein, Gordon (IEEE, 2015-06-07)
    • Defocus Deblurring and Superresolution for Time-of-Flight Depth Cameras

      Xiao, Lei; Heide, Felix; O'Toole, Matthew; Kolb, Andreas; Hullin, Matthias B.; Kutulakos, Kyros; Heidrich, Wolfgang (IEEE, 2015-06-07)
      Continuous-wave time-of-flight (ToF) cameras show great promise as low-cost depth image sensors in mobile applications. However, they also suffer from several challenges, including limited illumination intensity, which mandates the use of large numerical aperture lenses, and thus results in a shallow depth of field, making it difficult to capture scenes with large variations in depth. Another shortcoming is the limited spatial resolution of currently available ToF sensors. In this paper we analyze the image formation model for blurred ToF images. By directly working with raw sensor measurements but regularizing the recovered depth and amplitude images, we are able to simultaneously deblur and super-resolve the output of ToF cameras. Our method outperforms existing methods on both synthetic and real datasets. In the future our algorithm should extend easily to cameras that do not follow the cosine model of continuous-wave sensors, as well as to multi-frequency or multi-phase imaging employed in more recent ToF cameras.
    • TrackingNet: A Large-Scale Dataset and Benchmark for Object Tracking in the Wild

      Müller, Matthias; Bibi, Adel Aamer; Giancola, Silvio; Al-Subaihi, Salman; Ghanem, Bernard (Springer International Publishing, 2018-10-05)
      Despite the numerous developments in object tracking, further improvement of current tracking algorithms is limited by small and mostly saturated datasets. As a matter of fact, data-hungry trackers based on deep-learning currently rely on object detection datasets due to the scarcity of dedicated large-scale tracking datasets. In this work, we present TrackingNet, the first large-scale dataset and benchmark for object tracking in the wild. We provide more than 30K videos with more than 14 million dense bounding box annotations. Our dataset covers a wide selection of object classes in broad and diverse context. By releasing such a large-scale dataset, we expect deep trackers to further improve and generalize. In addition, we introduce a new benchmark composed of 500 novel videos, modeled with a distribution similar to our training dataset. By sequestering the annotation of the test set and providing an online evaluation server, we provide a fair benchmark for future development of object trackers. Deep trackers fine-tuned on a fraction of our dataset improve their performance by up to 1.6% on OTB100 and up to 1.7% on TrackingNet Test. We provide an extensive benchmark on TrackingNet by evaluating more than 20 trackers. Our results suggest that object tracking in the wild is far from being solved.
    • Experimental and Theoretical Study of Two-to-One Internal Resonance of MEMS Resonators

      Hajjaj, Amal Z.; Alfosail, Feras; Younis, Mohammad I. (ASME, 2018-11-02)
      In this paper, we investigate experimentally and theoretically the two-to-one (2:1) internal resonance between the first two symmetric vibrational modes of microelectromechanical (MEMS) arch resonator electrothermally tuned and electrostatically driven. Applying electrothermal voltage across the beam anchors controls its stiffness and then its resonance frequencies. Hence the ratio between the two frequencies can be tuned to a ratio of two. Then, we study the dynamic response of the arch beam during internal resonance. In the studied case, the presence of high AC bias excitation leads to the direct simultaneous excitation of the 1st and 3rd frequencies in addition to the activation of the internal resonance. A reduced order model and perturbation techniques are presented to analyze the nonlinear response of the structure. In the perturbation technique, the direct excitation of the 3rd resonance frequency is taken into consideration. Results show the presence of Hopf bifurcations, which can lead to chaotic motion at higher excitation. A good agreement among the theoretical and experimental results is shown.
    • Aerial Data Aggregation in IoT Networks: Hovering & Traveling Time Dilemma

      Bushnaq, Osama M.; Celik, Abdulkadir; ElSawy, Hesham; Al-Naffouri, Tareq Y.; Alouini, Mohamed-Slim (IEEE, 2018)
      The next era of information revolution will rely on aggregating big data from massive numbers of devices that are widely scattered in our environment. The majority of these devices are expected to be of low-complexity, low-cost, and limited power supply, which imposes stringent constraints on the network operation. In this regards, this paper proposes an aerial data aggregation from a finite spatial field via an unmanned aerial vehicle (UAV). Instead of fusing, relaying, and routing the data across the wireless nodes to fixed locations access points, a UAV flies over the field and collects the required data. Particularly, the field is divided into several subregions where the UAV hover over each subregion to collect samples from the underlying nodes. To this end, an optimization problem is formulated and solved to find the optimal number of subregions, the area of each subregion, the hovering locations, the hovering time at each location, and the trajectory between hovering locations such that an average number of samples are collected from the field in minimal time. The proposed the formulation is shown to be np-hard mixed integer problem, and hence, a decoupled heuristic solution is proposed. The results show that there exists an optimal number of subregions that balance the tradeoff between the hovering and traveling times such that the total time for collecting the required samples is minimized.
    • Megapixel Adaptive Optics: Towards Correcting Large-scale Distortions in Computational Cameras

      Wang, Congli; Fu, Qiang; Dun, Xiong; Heidrich, Wolfgang (Association for Computing Machinery (ACM), 2018-07)
      Adaptive optics has become a valuable tool for correcting minor optical aberrations in applications such as astronomy and microscopy. However, due to the limited resolution of both the wavefront sensing and the wavefront correction hardware, it has so far not been feasible to use adaptive optics for correcting large-scale waveform deformations that occur naturally in regular photography and other imaging applications. In this work, we demonstrate an adaptive optics system for regular cameras. We achieve a significant improvement in focus for large wavefront distortions by improving upon a recently developed high resolution coded wavefront sensor, and combining it with a spatial phase modulator to create a megapixel adaptive optics system with unprecedented capability to sense and correct large distortions.
    • Deep End-to-End Time-of-Flight Imaging

      Su, Shuochen; Heide, Felix; Wetzstein, Gordon; Heidrich, Wolfgang (2018)
      We present an end-to-end image processing framework for time-of-flight (ToF) cameras. Existing ToF image processing pipelines consist of a sequence of operations including modulated exposures, denoising, phase unwrapping and multipath interference correction. While this cascaded modular design offers several benefits, such as closed-form solutions and power-efficient processing, it also suffers from error accumulation and information loss as each module can only observe the output from its direct predecessor, resulting in erroneous depth estimates. We depart from a conventional pipeline model and propose a deep convolutional neural network architecture that recovers scene depth directly from dual-frequency, raw ToF correlation measurements. To train this network, we simulate ToF images for a variety of scenes using a time-resolved renderer, devise depth-specific losses, and apply normalization and augmentation strategies to generalize this model to real captures. We demonstrate that the proposed network can efficiently exploit the spatio-temporal structures of ToF frequency measurements, and validate the performance of the joint multipath removal, denoising and phase unwrapping method on a wide range of challenging scenes.
    • Super-Resolution and Sparse View CT Reconstruction

      Zang, Guangming; Aly, Mohamed; Idoughi, Ramzi; Wonka, Peter; Heidrich, Wolfgang (2018-09-01)
      We present a flexible framework for robust computed tomography (CT) reconstruction with a specific emphasis on recovering thin 1D and 2D manifolds embedded in 3D volumes. To reconstruct such structures at resolutions below the Nyquist limit of the CT image sensor, we devise a new 3D structure tensor prior, which can be incorporated as a regularizer into more traditional proximal optimization methods for CT reconstruction. As a second, smaller contribution, we also show that when using such a proximal reconstruction framework, it is beneficial to employ the Simultaneous Algebraic Reconstruction Technique (SART) instead of the commonly used Conjugate Gradient (CG) method in the solution of the data term proximal operator. We show empirically that CG often does not converge to the global optimum for tomography problem even though the underlying problem is convex. We demonstrate that using SART provides better reconstruction results in sparse-view settings using fewer projection images. We provide extensive experimental results for both contributions on both simulated and real data. Moreover, our code will also be made publicly available.
    • Depth and Transient Imaging with Compressive SPAD Array Cameras

      Sun, Qilin; Dun, Xiong; Peng, Yifan; Heidrich, Wolfgang (IEEE, 2018-06)
      Time-of-flight depth imaging and transient imaging are two imaging modalities that have recently received a lot of interest. Despite much research, existing hardware systems are limited either in terms of temporal resolution or are prohibitively expensive. Arrays of Single Photon Avalanche Diodes (SPADs) promise to fill this gap by providing higher temporal resolution at an affordable cost. Unfortunately SPAD arrays are to date only available in relatively small resolutions. In this work we aim to overcome the spatial resolution limit of SPAD arrays by employing a compressive sensing camera design. Using a DMD and custom optics, we achieve an image resolution of up to 800×400 on SPAD Arrays of resolution 64×32. Using our new data fitting model for the time histograms, we suppress the noise while abstracting the phase and amplitude information, so as to realize a temporal resolution of a few tens of picoseconds.
    • REST: A Reference-based Framework for Spatio-temporal Trajectory Compression

      Zhao, Yan; Shang, Shuo; Wang, Yu; Zheng, Bolong; Nguyen, Quoc Viet Hung; Zheng, Kai (ACM Press, 2018-07-19)
      The pervasiveness of GPS-enabled devices and wireless communication technologies results in massive trajectory data, incurring expensive cost for storage, transmission, and query processing. To relieve this problem, in this paper we propose a novel framework for compressing trajectory data, REST (Reference-based Spatio-temporal trajectory compression), by which a raw trajectory is represented by concatenation of a series of historical (sub-)trajectories (called reference trajectories) that form the compressed trajectory within a given spatio-temporal deviation threshold. In order to construct a reference trajectory set that can most benefit the subsequent compression, we propose three kinds of techniques to select reference trajectories wisely from a large dataset such that the resulting reference set is more compact yet covering most footprints of trajectories in the area of interest. To address the computational issue caused by the large number of combinations of reference trajectories that may exist for resembling a given trajectory, we propose efficient greedy algorithms that run in the blink of an eye and dynamic programming algorithms that can achieve the optimal compression ratio. Compared to existing work on trajectory compression, our framework has few assumptions about data such as moving within a road network or moving with constant direction and speed, and better compression performance with fairly small spatio-temporal loss. Extensive experiments on a real taxi trajectory dataset demonstrate the superiority of our framework over existing representative approaches in terms of both compression ratio and efficiency.
    • High Power GaN-based Blue Superluminescent Diode Exceeding 450 mW

      Alatawi, Abdullah; Holguin Lerma, Jorge Alberto; Shen, Chao; Shakfa, Mohammad Khaled; Alhamoud, Abdullah A.; Albadri, Abdulrahman M.; Alyamani, Ahmed Y.; Ng, Tien Khee; Ooi, Boon S. (IEEE, 2018-09-19)
      We demonstrate a high-power blue emitting superluminescent diode (SLD) with a tilted-facet configuration. An optical power of 457 mW with a broad spectral bandwidth of 6.5 nm was obtained under pulsed current injection of 1A, leading to a large power-bandwidth product of ~2970 mW·nm.
    • Analyzing Non-Orthogonal Multiple Access (NOMA) in Downlink Poisson Cellular Networks

      Ali, Konpal Shaukat; ElSawy, Hesham; Chaaban, Anas; Haenggi, Martin; Alouini, Mohamed-Slim (2018)
    • On the Optimization of Multi-Cell SLIPT Systems

      Abdelhady, Amr Mohamed Abdelaziz; Amin, Osama; Shihada, Basem; Alouini, Mohamed-Slim (2018)
      In this paper, we study the performance of simultaneous lightwave information and power transfer (SLIPT) systems of multi-cell indoor scenario. We aim to investigate the energy harvesting and data rate performance of multiple users while meeting the lightning constraints. To this end, we develop optimization frameworks and tune the light emitting diodes average currents to improve the performance of the SLIPT system. Firstly, we propose an algorithm to maximize the spectral efficiency (SE) subject to lighting and minimum harvested energy per user requirements. The proposed algorithm can be implemented in a distributed fashion with a reduced computational burden at each node. Then, we consider the energy harvesting maximization problem to investigate the maximum possible energy gain and its corresponding SE performance. Finally, we present some extensive simulations to explore the benefit of the optimization frameworks with respect to standard equal allocation setting. In addition, we monitor the effect of changing several system parameters on the two objectives and highlight the underlying trade-off between them.
    • Distributed In-Memory Analytics for Big Temporal Data

      Yao, Bin; Zhang, Wei; Wang, Zhi-Jie; Chen, Zhongpu; Shang, Shuo; Zheng, Kai; Guo, Minyi (Springer International Publishing, 2018-05-12)
      The temporal data is ubiquitous, and massive amount of temporal data is generated nowadays. Management of big temporal data is important yet challenging. Processing big temporal data using a distributed system is a desired choice. However, existing distributed systems/methods either cannot support native queries, or are disk-based solutions, which could not well satisfy the requirements of high throughput and low latency. To alleviate this issue, this paper proposes an In-memory based Two-level Index Solution in Spark (ITISS) for processing big temporal data. The framework of our system is easy to understand and implement, but without loss of efficiency. We conduct extensive experiments to verify the performance of our solution. Experimental results based on both real and synthetic datasets consistently demonstrate that our solution is efficient and competitive.
    • Human Supervised Multirotor UAV System Design for Inspection Applications

      Shaqura, Mohammad; Alzuhair, Khalid; Abdellatif, Fadl; Shamma, Jeff S. (IEEE, 2018-09-20)
      Multirotor UAVs are widely used for aerial inspection applications where missions are accomplished either via manual or autonomous control. Human controlled UAVs require trained pilots which can be a barrier from using the technology for general inspection personnel. Fully autonomous navigation, which employs onboard sensing, planning and coverage algorithms, is effective but comes with the cost of development and operational complexities. A human supervised UAV system design is presented where a deployed aerial vehicle operates in semi-autonomous mode. An operator, who is equipped with a smart handheld laser pointer, gives the UAV global guiding directions to reach the inspection target. The UAV is equipped with onboard vision sensing for local planning and target identification in addition to video streaming or recording. System operation is validated in indoor fliaht tests.