Search

Most of the existing saliency prediction research focuses on either single images or videos (or more precisely multiple images in sequence). However, to apply saliency prediction to drone exploration that has to consider multiple images from different view angles or localizations to determine the direction to explore, saliency prediction over multiple discontinuous images is required. In this paper, we propose a deep relative saliency model (MS-Net) for such an application. MS-Net starts with a single-image saliency feature extraction network for each image separately and then integrate these images by using a GCN-based mechanism called multi-image saliency fusion that learns relative saliency information among all the images. Finally, it predicts the saliency of each image by considering the relative information. Because there are no existing saliency prediction datasets with such multiple discontinuous images, we randomly cropped a large number of sub-images from 360° images of the existing 360° image saliency datasets to build our own dataset for both training and evaluation. Experimental results showed that the proposed MSNet considerably outperformed both single-image and video saliency prediction methods and could achieve comparative performance to that of 360° image saliency prediction even with only limited field-of-views, i.e., five sub-images, considered.

This paper will demonstrate the concept of a new, low-cost, on-line technique for monitoring battery state of health (SOH) and state of charge (SOC) using electrochemical impedance spectroscopy (EIS). A particular focus will be electric vehicles (EVs), where the SOC accuracy over existing battery management systems (BMS) will improve range prediction accuracy, although the proposed technique is also applicable to other electrochemical energy storage devices. While currently there exist few methods to measure the battery state of charge online, these methods are generally categorized as “indirect” methods which are prone to errors due to environmental changes and require additional hardware/costs for implementation. In this paper the EIS excitation signal will be generated by the system's existing power converter without requiring extra hardware but only requires software upgrade. The main objective of the proposed method is to minimize the impact on the main operation of the power converter in the system.

In this paper, we propose a novel, egocentric pointing system based on Google Glass which is equipped with an optical head-mounted display (OHMD) and a near-eye camera, with the eye-pointing line passing through the lower left corner of the display. For a pointed target, the pointing (or ranging) algorithm is based on a distance-pixel curve established from the camera-eye (epipolar) geometry. Additional pointing algorithms for estimating gazing point on a planar surface are also developed by establishing another distance-pixel curve along the same epipolar line. Experiments show that less than 0.32° angular error in the egocentric pointing can be achieved for object distance ranging from 80cm to 178cm by the best estimation scheme, with slightly less accurate results (i.e. 0.58°) achievable by simpler estimation schemes.

Rapid development of semiconductor technology and multi-function demands of end products has driven IC foundry industry toward 7nm node process, and even next generation of 5nm. The I/O pitch of chip is reduced accordingly but the build-up layer of IC carrier is still too large to fit interconnects. In order to overcome the gap of I/O pitch between IC chip and carrier, the interposer technology has been considered as a solution to resolve the issue. However, the cost of silicon interposer is too high, and the glass interposer lacks the associated infrastructure and is difficult to be handled, which makes a technology drawback for market applications. Alternatively, fan-out wafer/panel level package technology is getting more attractions for advanced package recently because of its features of low profile, small form factor, and high bandwidth with fine line re-distribution layer (RDL) routability. There are lots of literatures addressing about the residual stress and warpage mostly on wafer level fan-out technology, especially for chip-first technology scheme. However, comprehensive study on the panel level fan-out is not mature yet. This paper investigates fundamental factors that impact the residual stress and warpage level of panel level fan-out package, such as metal layer counts, thickness of dielectric and metal layer, coefficient of thermal expansion (CTE) and Young's modulus of dielectric and molding compound, molding gap and molding process temperature, etc. In this study, a RDL-first (chip-last) fan-out panel level structure of three metal layers on releasing film molded with epoxy compound was established as a simulation model by means of finite element analysis software. The simulation results provide a guideline of design rules for fabricating multi-layer RDL panel level fan-out package and making the minimum residual stress while chip assembly. Fabrication of three-layer dielectric panel level fan-out, where 370mmx470mm panel size is applied, is also demonstrated to compare with the simulation results.

Many applications of machine-to-machine (M2M) based intelligent transportation systems highly rely on the accurate estimation of neighbor map, where neighbor map mentions the locations of all nearby vehicles and pedestrians. To build the neighbor map, it usually integrates multiple sensors, such as GPS, odometer, inertial measurement unit (IMU), laser scanners, cameras, and RGB-D cameras. In this paper, we build a M2M framework to estimate the neighbor map and focus on the improvement of vehicle and pedestrian detection of most popular sensors, camera. We propose a novel grid-based object detection approach and deal with cameras on both roadside units and vehicles. It adapts to the environments and achieves high accuracy, and can be used to improve the performance of neighbor map estimation.

In this work, we develop an appearance-based gaze tracking system allowing user to move their head freely. The main difficulty of the appearance-based gaze tracking method is that the eye appearance is sensitive to head orientation. To overcome the difficulty, we propose a 3-D gaze tracking method combining head pose tracking and appearance-based gaze estimation. We use a random forest approach to model the neighbor structure of the joint head pose and eye appearance space, and efficiently select neighbors from the collected high dimensional data set. Li-optimization is then used to seek for the best solution for regression from the selected neighboring samples. Experiment results shows that it can provide robust binocular gaze tracking results with less constraints but still provides moderate estimation accuracy of gaze estimation.

The purpose of this study is to develop a non-invasive sleep monitoring system to distinguish sleep disturbances based on multiple sensors. Unlike clinical sleep monitoring which records biological information such as EEG, EOG, and EMG, in this study, we aim to identify occurrences of events from a sleep environment. A device with an infrared depth sensor, a RGB camera, and a four-microphone array is used to detect three types of events: motion events, lighting events, and sound events. Given streams of depth signals and color images, we build two background models to detect movements and lighting effects, and audio signals are scored simultaneously. Moreover, we classify events by an epoch approach algorithm and provide a graphical sleep diagram for browsing corresponding video clips. Experimental results in sleep condition show the efficiency and reliability of our system, and it is convenient and cost-effective to be used in home context.

Nowadays, ever expanding camera network makes it difficult to find the suspect from lengthy video records. This paper proposes a target-driven video summarization framework which provides two-step Filtered Summarized Video (FSV) for tracing suspects. Before the target is identified, users can find the target efficiently using the firststep FSV of any arbitrary camera. The first-step FSV filters all the attributes of the target including the time information and the target's categories. After identifying the target, the second-step FSV with additional spatio-temporal & appearance cues are triggered in the neighbor cameras. To enhance the accuracy of the object classification for FSV, we propose a Perspective Dependent Model (PDM) which consists of many grid-based models. Finally, the experimental results show that grid-based model is more robust than general detectors and the user study demonstrates better performance for target finding and tracking in camera network for surveillance.

We propose a tampering detection method using two-stage scene matching for real application with high efficiency and low false alarm rate. In the first stage, we use the intensity of edges as the main cue to detect the camera tampering events. Instead of using the entire edge points of the images, we sample the most significant edge points to represent the scene. Analyzing the edge variation with only the sample points, we discover that the events of camera tampering can be detected with low computation cost. Whenever the first stage detects the tampering event, the second stage is triggered to reduce false alarms. In the second stage, we propose an illumination change detector which can check the consistency of the scene structure using cell-based matching method. The experimental results demonstrate that our system can detect the camera tampering precisely and minimize false alarm even when the illumination changes dramatically or large crowds passing through the scene.

This paper presents a novel fully-integrated plastic-antibody-based biochip with easy-to-attach vacuum source for C-reactive protein sensing. Micropumps are the most popular devices for the microfluidic system especially in point-of-care devices. However, because the difficulty of miniaturizing the pump, most micropumps are complicated in the structure/assembly or not suitable for the disposable devices. In this paper, a fully-integrated biochip including easy-to-attach vacuum source made of shape memory polymer and a plastic antibody film were developed. By the 2 sets of vacuum source and mini-heater, the biochip could achieve two-stage microfluidic control with easy process. The vacuum sources show great performance in the experiment, and with the advantage of compact, simple structure, and east-to-attach with the biochip. The flow rate driven by our vacuum source is up to 15 μl/ml.

For target tracking across multiple cameras with disjoint views, previous works usually employed multiple cues and focused on learning a better matching model of each cue, separately. However, none of them had discussed how to integrate these cues to improve performance, to our best knowledge. In this paper, we look into the multi-cue integration problem and propose an unsupervised learning method since a complicated training phase is not always viable. In the experiments, we evaluate several types of score fusion methods and show that our approach learns well and can be applied to large camera networks more easily.

This paper proposes an adaptive learning method for tracking targets across multiple cameras with disjoint views. Two visual cues are usually employed for tracking targets across cameras: spatio-temporal cue and appearance cue. To learn the relationships among cameras, traditional methods used batch-learning procedures or hand-labeled correspondence, which can work well only within a short period of time. In this paper, we propose an unsupervised method which learns both spatio-temporal relationships and appearance relationships adaptively and can be applied to long-term monitoring. Our method performs target tracking across multiple cameras while also considering the environment changes, such as sudden lighting changes. Also, we improve the estimation of spatio-temporal relationships by using the prior knowledge of camera network topology.

Camera networks are often used in visual surveillance systems for wide-range monitoring. In this paper, we present a novel method for calibrating a camera network, which uses the trajectory of a bouncing ball as the calibration data. An important feature of our method is the use of the parabolic property of a ball's bouncing trajectory. This parabolic trajectory lies on a plane, called the parabolic trajectory plane (PT-plane), so that the relationship between the trajectory's points and their corresponding image points is a homography. Combining the vertical velocity determined by the earth's gravity and the horizontal velocity calculated from the homography, we can compute the 2D coordinates of the trajectory points on the PT-plane. By throwing the ball multiple times, we obtain calibration points on multiple planes for calibrating both intrinsic and extrinsic parameters of the networked cameras. Experimental results have demonstrated the feasibility and accuracy of the proposed method.