With the rapid development of Internet of Things (IoT) technologies, autonomous driving has become an integral part of the IoT ecosystem, where millimeter-wave radar plays a crucial role in ensuring robust perception under challenging weather and lighting conditions. However, its sparse and noisy data often require enhancement using high-end sensors such as LiDAR or RTK-GNSS, which are not common in commercial vehicles. This paper introduces mmEMP+, a self-supervised learning technique that leverages pervasive visual and inertial (VI) measurements to enhance radar sensing data. Using VI data to improve radar sensing introduces several challenges. First, moving objects in a scene are inaccurately reconstructed by VI structure-from-motion, which consequently fails to enhance radar sensing. Second, multipath effects generate spurious radar points that can distort the representation of the environment. Finally, the temporal misalignment between the camera, IMU, and mmWave radar results in mismatched data association, thereby degrading system performance. To address these issues, mmEMP+ first proposes a dynamic 3D reconstruction method to recover the positions of moving features accurately. Then, we develop a spatial-stability checking method to filter out spurious radar points. Finally, mmEMP+ devises a tightly coupled sensor fusion method to calibrate the multi-sensor temporal offset. Experiments on a real-world dataset show that mmEMP+ achieves performance comparable to high-channel LiDAR-supervised methods while using only low-cost sensors. We further validate its effectiveness in IoT-relevant applications such as object detection, localization, and mapping.

This article introduces a radar perception learning framework guided by data collected from commonly equipped visual-inertial (VI) sensor suites on smart vehicles. Unlike existing approaches that rely on dense point clouds from 3D LiDARs, which are costly and not widely deployed, this method leverages the broader availability of VI data. However, visual images alone lack the ability to capture the three-dimensional motion of moving targets, which limits their effectiveness in supervising motion-related tasks. To overcome this limitation, the framework integrates multiple perception tasks such as odometry estimation, motion segmentation, and scene flow prediction into a unified learning process. The first component is an odometry estimation module that combines deterministic ego-motion models with data-driven learning results. This fusion helps accurately infer the scene flow of static background points while minimizing drift. The second component is a supervision signal extraction module that aligns optical and millimeter-wave radar measurements to guide the learning of radar scene flow and rigid transformations. This module improves the reliability of dynamic point supervision through joint constraints across sensing modalities. The third component introduces a feature-selection module designed for cross-modal learning. It enhances the accuracy of motion segmentation and enforces consistency between odometry and scene flow, resulting in more coherent radar perception outputs. Experimental evaluations show that this framework achieves superior performance in challenging conditions such as smoke-obscured environments. It surpasses state-of-the-art (SOTA) methods that depend on high-cost LiDAR systems. The implementation of VISC+ will be open-source athttps://github.com/weini-Eve/VISC

This work proposes a mmWave radar’s scene flow estimation framework supervised by data from a widespread visual-inertial (VI) sensor suite, allowing crowdsourced training data from smart vehicles. Current scene flow estimation methods for mmWave radar are typically supervised by dense point clouds from 3D LiDARs, which are expensive and not widely available in smart vehicles. While VI data are more accessible, visual images alone cannot capture the 3D motions of moving objects, making it difficult to supervise their scene flow. Moreover, the temporal drift of VI rigid transformation also degenerates the scene flow estimation of static points. To address these challenges, we propose a drift-free rigid transformation estimator that fuses kinematic model-based ego-motions with neural network-learned results. It provides strong supervision signals to radar-based rigid transformation and infers the scene flow of static points. Then, we develop an optical-mmWave supervision extraction module that extracts the supervision signals of radar rigid transformation and scene flow. It strengthens the supervision by learning the scene flow of dynamic points with the joint constraints of optical and mmWave radar measurements. Extensive experiments demonstrate that, in smoke-filled environments, our method even outperforms state-of-the-art (SOTA) approaches using costly LiDARs.

System logs are crucial sources of information for engineers to analyze and resolve anomalies and faults in large-scale software systems. However, logs on a distributed system are often fragmented, making it challenging to achieve unified processing and comprehension. Traditional methods for log-based anomaly detection often employ machine learning algorithms with a focus on log event counts or log sequences. However, traditional methods fall short of fully leveraging the temporal and spatial structures inherent in distributed system logs, leading to issues of false positives and unstable performance in anomaly detection. In this paper, we propose a novel log anomaly detection method based on the construction of distributed system workflow graphs. This method extracts spatio-temporal information from distributed system logs and constructs event workflow graphs. These graphs accurately reflect the execution of the system and provide more comprehensive support for anomaly detection based on distributed system logs. The experimental results demonstrated that STGraph achieved F1 scores of 0.959,0.979, and 0.959 on HDFS, BGL, and OpenStack datasets respectively, outperforming LogRobust, PLELog, and NeuralLog by 1.2%-18.6% across precision/recall metrics. Notably, it attained 0.985 recall on BGL and maintained >0.935 F1 scores under 30% noise interference, 21.8% higher than LogRobust.

Polar sea ice monitoring is critical for global climate research and Arctic ship navigation, yet traditional satellite remote sensing struggles to resolve micro-scale ice features due to resolution and weather constraints. This study presents a novel sea ice classification method integrating UAV-derived high-resolution optical imagery from the 2024 “Sun Yat-sen University Polar Vessel” Arctic expedition with deep learning. The dataset, comprising four types (sea ice, melt ponds, open water, ships), was constructed through geometric correction, and semi-automated annotation using the ISAT-SAM approach. The proposed Ice-Unet model, optimized via transfer learning from pre-trained weights, employs a VGG16-inspired backbone with multi-scale feature fusion and downscaling mechanisms for semantic segmentation. Comparative evaluations against DeepLabV3+ (Xception/MobileNet), PSPNet (ResNet50), and HRNet demonstrate Ice-Unet’s superiority, achieving leading IoU of 95.07% and recall rates of 97.11%, particularly excelling in melt pond detection and ice lead morphology characterization. Case analysis reveals precise quantification of melt pond spatial patterns and predictive capability for their evolution into sea ice leads. The research empirically confirms the operational synergy between UAV platforms and deep learning architectures in monitoring transient sea ice dynamics, establishing a sub-meter resolution methodology that significantly enhances the precision of sea ice classification based ship navigation through near-real-time geospatial analytics.

Large Language Models (LLMs), with their strong generalization and inference capabilities, have been increasingly leveraged to address the challenges of handling corner cases in autonomous driving (AD). However, a critical unresolved issue remains: the lack of a comprehensive understanding and formal assessment of LLMs’ driving theory knowledge and practical skills. To address this issue, we propose the first dedicated driving theory test framework and benchmark for LLMs. That is a crucial yet unexplored area in the literature, particularly for safety-critical applications in autonomous driving and driver assistance. Our framework systematically evaluates LLMs’ competence in driving theory and hazard perception, akin to the official UK driving theory test, ensuring their qualification for critical driving-related tasks. To facilitate rigorous benchmarking, we construct a comprehensive dataset comprising over 700 multiple-choice questions (MCQs) and 54 hazard perception video tests sourced from the official UK driving theory examination. Additionally, we incorporate two standardized MCQ sets from the UK’s Driver and Vehicle Standards Agency (DVSA). For these two types of theoretical test items, we design tailored assessment methodologies and evaluation metrics, including accuracy, recall, precision, F1-score, real-time performance, and computational efficiency. The experimental results reveal that among all LLMs tested, only GPT-4o achieved an accuracy of 88. 21% in the MCQs test, successfully passing this component. However, in hazard perception testing, none of the evaluated models met the passing criteria under the given settings, highlighting the substantial improvements required before these models can be practically deployed for real-world driving applications. Our key insight is that the specific test questions LLMs fail to answer correctly directly reflect their deficiencies in understanding and flexibly applying traffic regulations, as well as in analyzing and responding to complex driving scenarios. This provides clear directions for future improvements.

Logs play a crucial role in recording valuable system runtime information, extensively utilized by service providers and users for effective service management. A typical approach in service management, based on log analysis, involves parsing the original log messages initially presented in an unstructured format. Subsequently, a data mining model is employed to extract critical system behavior information, aiding in service management. As the volume of logs rapidly increases, training models using current log resolution methods post-log collection becomes excessively time-consuming, leading to decreased accuracy. Manual analysis of extensive logs is both time-intensive and inefficient. This article introduces Aclog, an automated log parsing tool tailored for large-scale log analysis, storage, and management. Aclog operates by storing and managing logs in a structured and unified format, thereby offering a cohesive database for comprehensive log auditing of computing systems. Key components of Aclog encompass the log updater, log parser, log storage, and log querier. In this paper, we utilize a realworld, large-scale public log dataset to showcase the capabilities of Aclog. We evaluate the log files generated by ten popular systems.

Complementary to prevalent LiDAR and camera systems, millimeter-wave (mmWave) radar is robust to adverse weather conditions like fog, rainstorms, and blizzards but offers sparse point clouds. Current techniques enhance the point cloud by the supervision of LiDAR’s data. However, high-performance LiDAR is notably expensive and is not commonly available on vehicles. This paper presents mmEMP, a supervised learning approach that enhances radar point clouds using a low-cost camera and an inertial measurement unit (IMU), enabling crowd-sourcing training data from commercial vehicles. Bringing the visual-inertial (VI) supervision is challenging due to the spatial agnostic of dynamic objects. Moreover, spurious radar points from the curse of RF multipath make robots misunderstand the scene. mmEMP first devises a dynamic 3D reconstruction algorithm that restores the 3D positions of dynamic features. Then, we design a neural network that densifies radar data and eliminates spurious radar points. We build a new dataset in the real world. Extensive experiments show that mmEMP achieves competitive performance compared with the SOTA approach training by LiDAR’s data. In addition, we use the enhanced point cloud to perform object detection, localization, and mapping to demonstrate mmEMP’s effectiveness.

In visual SLAM (VSLAM) systems, loop closure plays a crucial role in reducing accumulated errors. However, VSLAM systems relying on low-level visual features often suffer from the problem of perceptual confusion in repetitive environments, where scenes in different locations are incorrectly identified as the same. Existing work has attempted to introduce object-level features or artificial landmarks. The former approach struggles to distinguish visually similar but different objects, while the latter is both time-consuming and labor-intensive. This paper introduces a novel loop closure detection method that leverages pretrained AI foundation models to extract rich semantic information about specific types of objects (e.g., door numbers), referred to as semantic anchors, that help to distinguish similar scenes better. In settings such as office buildings, hotels, and warehouses, this approach helps to improve the robustness of loop closure detection. We validate the effectiveness of our method through experiments conducted in both simulated and real-world environments.

Edge-assisted visual SLAM stands as a pivotal enabler for emerging mobile applications, such as search-and-rescue, smart logistics, and industrial inspection. Limited by the computing capability of lightweight mobile devices like MAVs, current innovations balance system accuracy and efficiency by allocating lightweight and time-sensitive tracking tasks to mobile devices, while offloading the more resource-intensive yet delay-tolerant map optimization tasks to the edge. However, our pilot study in a large-scale oil field reveals several limitations of such a tracking-optimization decoupled paradigm, arising due to the disruption of inter-dependencies between the two tasks concerning data, resources, and threads.In this paper, we design and implement edgeSLAM2, an innovative system that reshapes the edge-assisted visual SLAM paradigm by tightly integrating tracking and partial-yet-crucial optimization on mobile. edgeSLAM2 harnesses the hierarchical and heterogeneous computing units offered by the latest commercial systems-on-chip (SoCs) to enhance the computational capacity of mobile devices, which in turn, allows edgeSLAM2 to design a suit of novel algorithms for map sync, optimization, and tracking that accommodate such architectural upgrade. By fully embracing the on-chip intelligence, edgeSLAM2 simultaneously enhances system accuracy and efficiency through software-hardware co-design. We deploy edgeSLAM2 on an industrial drone and conduct comprehensive experiments in a large-scale oil field over three months. The results show that edgeSLAM2 surpasses comparative methods by achieving an 80% reduction in bandwidth consumption, a 32% improvement in accuracy, and a 26% reduction in tracking delay.

Constructing reliable trajectories for ship targets in busy waterways involves associating tracks from various sensor sources. This poses challenges due to the incoherence in sensor data precision and acquisition frequency, the high density of targets, and the intertwined correlation scenarios involving trajectory interruptions and multi-source (also known as multi-sensor) observations. Utilizing the intuitive idea that a ship's motion pattern is influenced by its geographic location, we integrate geographic and dynamic information to enhance trajectory accuracy in dense scenes. The proposed approach employs self-supervised contrastive learning to extract relevant features, accompanied by a dual-level augmentation strategy tailored to two association scenarios. Additionally, a specialized feature enhancement module is introduced to enable a comprehensive representation of trajectories. When tested on a large dataset and compared with established methods, our method demonstrates superior capabilities in extracting trajectory features and excels in a variety of association contexts.

Metric localization is vital to autonomous driving where it corrects cumulative errors in a long-term run. Such errors are inevitable in real scenarios where GPS signals or some other drift-free exteroceptive measurements are not available, e.g., when an automobile goes through a tunnel. Using FMCW-based mmWave radars is an attractive metric localization technique with improved robustness as RF signals can traverse small particles in harsh weather conditions like snowing, foggy, and storming, but it faces a fundamental challenge of Doppler distortion. Existing works take spatial constraints to mitigate the Doppler distortion of point clouds from mechanical radars with limited accuracy. Modern single-chip mmWave radars that provide dynamic estimates, i.e., radial velocities, bring new opportunities to develop more accurate approaches. This paper presentsDC-Loc++, a robust metric localization framework by compensating Doppler distortions using a single-chip mmWave radar. It consists of an explicit velocity-assisted Doppler compensation module for each radar sub-map, an uncertainty-aware metric registration algorithm, and a failure recovery method that validates measurement constraints to generate a more confident pose graph for optimizing vehicle poses. Extensive experiments on both nuScenes dataset and a synthetic CARLA dataset show the effectiveness ofDC-Loc++, achieving 99.2% success rate and more than 20.0%, 30.2% error reductions in terms of translation and rotation estimates, respectively, compared with existing approaches.

Edge-assisted visual SLAM plays a crucial role in enabling innovative mobile applications, such as autonomous swarm inspection, search-and-rescue, and smart logistics. Constrained by the computational capacities of lightweight mobile devices, current approaches delegate lightweight, time-sensitive tracking tasks to the mobile end while offloading resource-intensive, latency-tolerant map optimization tasks to the edge. However, our pilot study reveals several limitations of the tracking-optimization decoupled paradigm, stemming from the disruption of inter-dependencies between the two tasks. In this paper, we design and implement edgeSLAM2, an innovative system that reshapes the edge-assisted visual SLAM paradigm by tightly integrating tracking and partial-yet-crucial optimization on mobile. edgeSLAM2 harnesses the heterogeneous computing units offered by the commercial systems-on-chip (SoCs) to enhance the computational capacity of mobile devices, which in turn, allows edgeSLAM2 to design a suit of novel algorithms for map sync, optimization, and tracking that accommodate such architectural upgrade. By capitalizing on the full potential of on-chip intelligence, edgeSLAM2 supports both solitary and collaborative SLAM with accuracy and immediacy, underpinned by a cohesive software-hardware co-design. We deploy edgeSLAM2 on drones for industrial inspection. Comprehensive experiments in one of the world’s largest oil fields over three months demonstrate its superior performance.

In this abstract, a decision-level fusion method by utilizing Synthetic Aperture Radar (SAR) and passive microwave remote sensing data for sea ice concentration estimation is investigated. In the proposed method, the merits from passive microwave data and SAR data are both taken into consideration. For SAR imagery, incident angles and azimuth angles were used to correct backscattering values from slant range to ground range in order to improve geocoding accuracy. Within a pixel, sea ice concentration from passive microwave data and sea ice category label derived from conditional random fields (CRF) framework in SAR imagery are calibrated under the least distance protocol. Then, posterior probability distribution between category label derived from SAR imagery and passive microwave sea ice concentration product is modeled and integrated under the Bayesian network. In the posterior probability estimation procedure, final sea ice concentration is obtained using maximum a posteriori probability (MAP) criterion, which equals to minimize the cost function by nonlinear iteration method. We construct the constrained least-squared method to derive sea ice concentration from passive microwave data. Sea ice type category is exploited by the proposed CRF based strategies including the mixed statistical conditional random fields (MSTA-CRF) and fully connected. In the experiments, results show that the proposed algorithm outperform both the concentration from SAR and the passive microwave product. Especially, the proposed fusion method can improve the accuracy of passive sea ice concentration products and reduce the uncertainty around the ice edge and thin ice area.

Security check-in life-support areas, e.g., bridge and engine room are crucial for cruise ships due to numerous and diverse passenger identities. Instead of conventional security check approaches, such as facial recognition and fingerprint identification, device-free approaches enabled by WiFi-based gait recognition have attracted considerable attention owing to their low cost, nonintrusiveness, and privacy protection. Despite the excellent performance of existing indoor methods, they cannot be trivially extended to cruise ships because of the unique characteristics of hull deformation caused by vibrating engines and waves. This stems from the flexible structure of cruise ships, which introduces additional noise to the WiFi signals. To address this challenge, we propose WiCrew, a device-free gait recognition system that detects crew identity anomalies in cruise ships. WiCrew consists of two components: 1) a spatial separation algorithm that separates the signal components from ship vibration and human activity and 2) a speed-independent adversarial learning framework that identifies the ship’s crew using human gaits at an arbitrary walking speed. Extensive experiments on a cruise ship demonstrate the effectiveness of WiCrew. While the crew members walk at speed of 0.7 to 1.8 m/s, the average recognition accuracy reaches 82%, which is similar to vision-based approaches.

Recent advances in generative adversarial networks (GANs) have provided potential solutions for photo-realistic human image synthesis. However, the explicit and individual control of synthesis over multiple factors, such as poses, body shapes, and skin colors, remains difficult for existing methods. This is because current methods mainly rely on a single pose/appearance model, which is limited in dis-entangling various poses and appearance in human images. In addition, such a unimodal strategy is prone to causing severe artifacts in the generated images like color distortions and unrealistic textures. To tackle these issues, this paper proposes a multi-factor conditioned method dubbed BodyGAN. Specifically, given a source image, our Body-GAN aims at capturing the characteristics of the human body from multiple aspects: (i) A pose encoding branch consisting of three hybrid subnetworks is adopted, to generate the semantic segmentation based representation, the 3D surface based representation, and the key point based rep-resentation of the human body, respectively. (ii) Based on the segmentation results, an appearance encoding branch is used to obtain the appearance information of the human body parts. (iii) The outputs of these two branches are represented by user-editable condition maps, which are then processed by a generator to predict the synthesized image. In this way, our BodyGAN can achieve the fine-grained dis-entanglement of pose, body shape, and appearance, and consequently enable the explicit and effective control of syn-thesis with diverse conditions. Extensive experiments on multiple datasets and a comprehensive user study show that our BodyGAN achieves the state-of-the-art performance.

Automotive mmWave radar has been widely used in the automotive industry due to its small size, low cost, and complementary advantages to optical sensors (e.g., cameras, LiDAR, etc.) in adverse weathers, e.g., fog, raining, and snowing. On the other side, its large wavelength also poses fundamental challenges to perceive the environment. Recent advances have made breakthroughs on its inherent drawbacks, i.e., the multipath reflection and the sparsity of mmWave radar's point clouds. However, the frequency-modulated continuous wave modulation of radar signals makes it more sensitive to vehicles’ mobility than optical sensors. This work focuses on the problem of frequency shift, i.e., the Doppler effect distorts the radar ranging measurements and its knock-on effect on metric localization. We propose a new radar-based metric localization framework, termed DC-Loc, which can obtain more accurate location estimation by restoring the Doppler distortion. Specifically, we first design a new algorithm that explicitly compensates the Doppler distortion of radar scans and then model the measurement uncertainty of the Doppler-compensated point cloud to further optimize the metric localization. Extensive experiments using the public nuScenes dataset and CARLA simulator demonstrate that our method outperforms the state-of-the-art approach by 25.2% and 5.6% improvements in terms of translation and rotation errors, respectively.

Abstract Environmental pollution is one of the major challenges in China, which seriously affects people's life, economy and society. This paper proposes a multi‐criteria group decision‐making method based on hesitant fuzzy preference relations with self‐confidence (HFPRs‐SC) behaviours and applies it to the evaluation of the environmental pollution emergency response process. Firstly, a comprehensive evaluation system is proposed that covers the environmental pollution emergency response process. Moreover, is utilized to obtain the normalized hesitant fuzzy preference relationship, and an algorithm that considers both the hesitant fuzzy preference value and the self‐confidence level is proposed to improve the consistency of HFPRs‐SC. In addition, combined with subjective and objective weights of experts, a weighted average operator based on the confidence level is applied to aggregate individual HFPR‐SC. Furthermore, the score function of HFPRs‐SC is designed to get the best ranking. Finally, a case study on an explosion accident in Fujian Province of China is conducted to verify the practicability and effectiveness of the proposed method with comparative analysis and sensitivity analysis. Based on the analysis, some suggestions are put forward for improving environmental pollution emergency response.

Being able to track passengers’ movement without invasion of their privacy plays an important role in cruise ships; it enables crucial location-based services, such as maritime search and rescue, tourist services, and epidemic prevention. The past few years have witnessed commodity WiFi holding great potential that provides such services available thanks to its ubiquitous in indoor scenarios. However, existing WiFi-based tracking methods suffer from huge performance degradation in sailing ships due to their complex metal structures and dynamic hull deformation caused by engines and waves/payloads pressure. In this article, we present CRLoc, a deep learning-based passive human tracking system that can overcome the practical limitations of traditional WiFi-based localization approaches applied in a multipath-rich and mobile ship environment, and provide decimeter-level tracking accuracy in cruise ships. Specifically, we make two contributions, i.e., we propose a super-resolution parameter estimation algorithm that better characterizes ship indoor environments, and a deep neural network-based end-to-end solution to remove the impact of noise, interference, and mobility in ships. The real-world implementation and extensive experiments in several passenger ships demonstrate that CRLoc tracks human motions with a median error of 92 cm, better than state-of-the-art localization methods. To our knowledge, this is one of the first WiFi-based passive human motion tracking system in a cruise ship environment.

Sea ice plays an important role in global climate system. Operation IceBridge (OIB) was launched in 2009 to bridge the gap in polar observations between the Ice, Cloud, and Land Elevation Satellite (ICESat) and ICESat-2 missions. OIB flew more than 1000 aircraft surveys, providing annual mapping of sea ice, glaciers and ice sheets in the Antarctic and Arctic. This study utilizes altimetry data and synchronized optical images from OIB to derive Arctic sea ice freeboard estimates for the period from 2009 to 2019. Lead detection is important in determining sea ice freeboard. We developed an improved approach by combining optical images, OIB laser pulse apparent reflectivity, and L1B elevation profiles to detect leads. We then determined the local sea surface height with a lowest elevation method from the L1B elevation profile of leads. Thus, sea ice freeboard was calculated by the difference between L2 relative elevation and corresponding local sea surface height. We derived the estimates of the Arctic sea ice freeboard from 2009 to 2019. Spatial and temporal variations of the freeboard were analyzed based on the estimates. The Arctic annual average sea ice freeboard showed a decreasing trend from OIB observations. However, an obvious increase in multiyear ice freeboard was observed in 2014. We validated our freeboard estimates with OIB freeboard product from the National Snow and Ice Data Center and coincident ICESat-2 freeboard. The comparison shows that our freeboard estimates generally compare well with the ICESat-2 freeboard, and our method is more sensitive to thin ice.

The advances in agile micro aerial vehicles (MAVs) have shown great potential in replacing humans for labor-intensive or dangerous indoor investigation, such as warehouse management and fire rescue. However, the design of a state estimation system that enables autonomous flight poses fundamental challenges in such dim or smoky environments. Current dominated computer-vision based solutions only work in well-lighted texture-rich environments. This paper addresses the challenge by proposing Marvel, an RF backscatter-based state estimation system with online initialization and calibration. Marvel is nonintrusive to commercial MAVs by attaching backscatter tags to their landing gears without internal hardware modifications, and works in a plug-and-play fashion with an automatic initialization module. Marvel is enabled by three new designs, a backscatter-based pose sensing module, an online initialization and calibration module, and a backscatter-inertial super-accuracy state estimation algorithm. We demonstrate our design by programming a commercial MAV to autonomously fly in different trajectories. The results show that Marvel supports navigation within a range of 50 m or through three concrete walls, with an accuracy of 34 cm for localization and 4.99for orientation estimation. We further demonstrate our online initialization and calibration by comparing to the perfect initial parameter measurements from burdensome manual operations.

Recovering dense human poses from images plays a critical role in establishing an image-to-surface correspondence between RGB images and the 3D surface of the human body, serving the foundation of rich real-world applications, such as virtual humans, monocular-to-3d reconstruction. However, the popular DensePose-COCO dataset relies on a sophisticated manual annotation system, leading to severe limitations in acquiring the denser and more accurate annotated pose resources. In this work, we introduce a new 3D human-body model with a series of decoupled parameters that could freely control the generation of the body. Furthermore, we build a data generation system based on this decoupling 3D model, and construct an ultra dense synthetic benchmark UltraPose, containing around 1.3 billion corresponding points. Compared to the existing manually annotated DensePose-COCO dataset, the synthetic UltraPose has ultra dense image-to-surface correspondences without annotation cost and error. Our proposed UltraPose provides the largest benchmark and data resources for lifting the model capability in predicting more accurate dense poses. To promote future researches in this field, we also propose a transformer-based method to model the dense correspondence between 2D and 3D worlds. The proposed model trained on synthetic UltraPose can be applied to real-world scenarios, indicating the effectiveness of our benchmark and model.1

The versatile nature of agile micro aerial vehicles (MAVs) poses fundamental challenges to the design of robust state estimation in various complex environments. Achieving high-quality performance in textureless scenes is one of the missing pieces in the puzzle. Previously proposed solutions either seek a remedy with visual loop closure or leverage RF localizability with inferior accuracy. None of them support accurate MAV state estimation in textureless scenes. This paper presents RFSift, a new state estimator that conquers the textureless challenge with RF-referenced monocular vision, achieving centimeter-level accuracy in textureless scenes. Our key observation is that RF and visual measurements are tied up with pose constraints. Mapping RF to feature quality and sift well-matched ones significantly improves accuracy. RFSift consists of 1) an RF-sifting algorithm that maps 3D UWB measurements to 2D visual features for sifting the best features; 2) an RF-visual-inertial sensor fusion algorithm that enables robust state estimation by leveraging multiple sensors with complementary advantages. We implement the prototype with off-the-shelf products and conduct large-scale experiments. The results demonstrate that RFSift is robust in textureless scenes, 10x more accurate than the state-of-the-art monocular vision system. The code of RFSift is available at https://github.com/weisgroup/RFSift.

Cloud-based Internet of Things, which is considered as a promising paradigm these days, can provide various applications for our society. However, as massive sensitive and private data in IoT devices are collected and outsourced to cloud for data storage, processing, or sharing for cost saving, the data security has become a bottleneck for its further development. Moreover, in many large-scale IoT systems, multiple group data sharing is practical for users. Thus, how to ensure data security in multiple group data sharing remains an open problem, especially the fine-grained access control and data integrity verification with public auditing. Therefore, in this paper, we propose a blockchain-based fine-grained data sharing scheme for multiple groups in cloud-based IoT systems. In particular, we design a novel multiauthority large universe CP-ABE scheme to guarantee the fine-grained access control and data integrity across multiple groups by integrating group signature into our scheme. Moreover, to ease the need for a trusted third auditor in traditional data public auditing schemes, we introduce blockchain technique to enable a distributed data public auditing. In addition, with the group signature, our scheme also realizes anonymity and traitor tracing. The security analysis and performance evaluation show that our scheme is practical for large-scale IoT systems.

he advances in compact and agile micro aerial vehicles (MAVs) have shown great potential in replacing human for labor-intensive or dangerous indoor investigation, such as warehouse management and fire rescue. However, the design of a state estimation system that enables autonomous flight in such dim or smoky environments presents a conundrum: conventional GPS or computer vision based solutions only work in outdoors or well-lighted texture-rich environments. This paper takes the first step to overcome this hurdle by proposing Marvel, a lightweight RF backscatter-based state estimation system for MAVs in indoors. Marvel is nonintrusive to commercial MAVs by attaching backscatter tags to their landing gears without internal hardware modifications, and works in a plug-and-play fashion that does not require any infrastructure deployment, pre-trained signatures, or even without knowing the controller's location. The enabling techniques are a new backscatter-based pose sensing module and a novel backscatter-inertial super-accuracy state estimation algorithm. We demonstrate our design by programming a commercial-off-the-shelf MAV to autonomously fly in different trajectories. The results show that Marvel supports navigation within a range of 50 m or through three brick walls, with an accuracy of 34 cm for localization and 4.99° for orientation estimation, outperforming commercial GPS-based approaches in outdoors.

Recent years have witnessed the rapid proliferation of backscatter technologies that realize the ubiquitous and long-term connectivity to empower smart cities and smart homes. Localizing such backscatter tags is crucial for IoT-based smart applications. However, current backscatter localization systems require prior knowledge of the site, either a map or landmarks with known positions, which is laborious for deployment. To empower universal localization service, this paper presents Rover, an indoor localization system that localizes multiple backscatter tags without any start-up cost using a robot equipped with inertial sensors. Rover runs in a joint optimization framework, fusing measurements from backscattered WiFi signals and inertial sensors to simultaneously estimate the locations of both the robot and the connected tags. Our design addresses practical issues including interference among multiple tags, real-time processing, as well as the data marginalization problem in dealing with degenerated motions. We prototype Rover using off-the-shelf WiFi chips and customized backscatter tags. Our experiments show that Rover achieves localization accuracies of 39.3 cm for the robot and 74.6 cm for the tags.

Recent years have witnessed the rapid proliferation of low- power backscatter technologies that realize the ubiquitous and long-term connectivity to empower smart cities and smart homes. Localizing such low-power backscatter tags is crucial for IoT-based smart services. However, current backscatter localization systems require prior knowledge of the site, either a map or landmarks with known positions, increasing the deployment cost. To empower universal localization service, this paper presents Rover, an indoor localization system that simultaneously localizes multiple backscatter tags with zero start-up cost using a robot equipped with inertial sensors. Rover runs in a joint optimization framework, fusing WiFi-based positioning measurements with inertial measurements to simultaneously estimate the locations of both the robot and the connected tags. Our design addresses practical issues such as the interference among multiple tags and the real- time processing for solving the SLAM problem. We prototype Rover using off-the-shelf WiFi chips and customized backscatter tags. Our experiments show that Rover achieves localization accuracies of 39.3 cm for the robot and 74.6 cm for the tags.

We present WINS, a state estimator that fuses commodity Wi-Fi and an inertial sensor (IMU) to enable indoor autonomous flight of MAVs. It overcomes the lighting and environmental texture limitations of current vision-based approaches. WINS incorporates two modules: First, a real-time AoA estimation algorithm that outputs drift-free measurements up to 20 Hz to confine the drift of IMU. Second, a novel WiFi-inertial state estimator to let our highly nonlinear system work without the need of prior initializations and without knowing the position of Wi-Fi infrastructure (APs). The preliminary results show that WINS achieves a mean MAV's location accuracy of 61.7 cm with a maximum flying velocity of 1.27 m/s.

Cluster scheduler serves as a critical component to data-parallel systems in datacenters. Ideally, a scheduler should provide predictable performance with guarantees on the maximal job completion delay, while at the same time ensuring the minimal mean response time. Practically however, performance predictability and optimality are often conflicting with each other. The results often are a plethora of scheduling policies that either achieve predictable performance at the expense of long response times (e.g., max-min fairness), or run the risk of starving some jobs to obtain the minimal mean response time (e.g., Shortest Remaining Processing Time First). To address these problems, we develop a new scheduler, Cluster Fair Queueing (CFQ), which preferentially offers resources to jobs that complete the earliest under a fair sharing policy. We show that CFQ is able to minimize the mean response time while at the same time ensuring jobs to finish within a constant time after their completion under fair sharing. Our Spark deployment on a 100-node EC2 cluster demonstrates that compared to the built-in fair scheduler, CFQ can decrease the mean response time by 40%, which speeds up more than 40% of jobs by over 75% on average.

In next-generation LTE-advanced cellular networks, device-to-device (D2D) communication has emerged as an effective way to offload cellular traffic and improve system performance. Conventionally, a device exclusively relies on cellular communication to retrieve the content it desires. With D2D communication, however, if the same piece of content is available in the vicinity of the device, the content can be directly retrieved from one of its neighbouring devices. Naturally, the key problem becomes how to maximize content sharing via D2D communication. Existing works on content sharing are mainly concerned with a multi-hop communication setting, while works on D2D communication have primarily focused on the communication aspects, including interference avoidance and energy efficiency. In this paper, we study the problem of maximizing cellular traffic offloading with D2D communication, by selectively caching popular content locally, and by exploring maximal matching for sender-receiver pairs. Specifically, we consider an interference-aware communication model and formulate selective caching as a Knapsack problem, and sender-receiver matching as a maximum weighted matching problem in a bipartite graph. We propose decentralized algorithms to solve both problems, and our simulation results demonstrate that our algorithms are effective in maximizing cellular traffic offloading.

Though HTTP adaptive video streaming has been widely adopted in the industry, it has been shown that it suffers from lackluster performance with respect to a number of desirable properties: fairness, efficiency, and stability, especially when multiple players compete for a bottleneck link. Effective algorithms have been proposed for single-server HTTP adaptive streaming to mitigate these problems. In this paper, we present an in-depth analysis on these desirable properties in the context of using HTTP adaptive streaming from multiple servers, which demonstrate that these properties will no longer hold with the existing algorithms. To address this challenge, we present Presto, a new protocol designed to improve the user experience by providing better fairness, efficiency and stability in the context of multi-server HTTP adaptive streaming. Our real-world experimental results indicate that Presto substantially outperforms existing protocols in the multiple server scenario.

Temporal tie points have been manually selected for sea ice concentration retrieval based on the linear combination relationship between the open water and the complete ice coverage pixel. In this paper, the multichannel information has been exploited using the constrained least-squares linear unmixing algorithm from AMSR-E multi-channels brightness temperature. Snow depth can be obtained from the innovative algorithm without considering the weather effect.

Sensor networks are invariably coupled tightly with the geometric environment in which the sensor nodes are deployed. Network boundary is one of the key features that characterize such environments. While significant advances have been made for 2D cases, so far boundary extraction for 3D sensor networks has not been thoroughly studied. We present CABET, a novel Connectivity-Based Boundary Extraction scheme for large-scale 3D sensor networks. To the best of our knowledge, CABET is the first 3D-capable and pure connectivity-based solution for detecting sensor network boundaries. It is fully distributed, and is highly scalable, requiring overall message cost linear with the network size. A highlight of CABET is its non-uniform critical node sampling , called r'-sampling , that selects landmarks to form boundary surfaces with bias toward nodes embodying salient topological features. Simulations show that CABET is able to extract a well-connected boundary in the presence of holes and shape variation, with performance superior to that of some state-of-the-art alternatives. In addition, we show how CABET benefits a range of sensor network applications including 3D skeleton extraction, 3D segmentation, and 3D localization.

As a sensor network grows large, it may become increasingly complex in topology due to its close ties to the surrounding environment. Previous work has shown that proper geometric processing of the network (e.g., boundary detection and localization) can provide very helpful information for applications to optimize their performance. To that end, numerous algorithms have been developed, providing a variety of inspiring solutions, yet exhibiting an ad hoc style in principle and implementation. In this paper we show that the crux of solving many of the problems caused by complex topology is to identify the concave nodes, nodes that are located at concave network corners, where the boundary has an inner angle greater than π. The knowledge of such nodes makes several important tasks, namely geometric embedding, full localization, convex segmentation, and boundary detection, relatively easier or perform significantly better, as confirmed by simulations. These findings suggest that concave nodes can serve as a basic supporting structure for general geometric processing tasks and geometry-related applications in sensor networks.

A connectivity-based and anchor-free three-dimensional localization (CATL) scheme is presented for large-scale sensor networks with concave regions. It distinguishes itself from previous work with a combination of three features: (1) it works for networks in both 2D and 3D spaces, possibly containing holes or concave regions; (2) it is anchor-free and uses only connectivity information to faithfully recover the original network topology, up to scaling and rotation; (3) it does not depend on the knowledge of network boundaries, which suits it well to situations where boundaries are difficult to identify. The key idea of CATL is to discover the notch nodes , where shortest paths bend and hop-count-based distance starts to significantly deviate from the true Euclidean distance. An iterative protocol is developed that uses a notch-avoiding multilateration mechanism to localize the network. Simulations show that CATL achieves accurate localization results with a moderate per-node message cost.

Sensor networks are invariably coupled tightly with the geometric environment in which the sensor nodes are deployed. Network boundary is one of the key features that characterize such environments. While significant advances have been made for 2D cases, so far boundary extraction for 3D sensor networks has not been thoroughly studied. We present CABET, a novel Connectivity-bAsed Boundary Extraction scheme for large-scale Three-dimensional sensor networks. To the best of our knowledge, CABET is the first 3D-capable and pure connectivity-based solution for detecting sensor network boundaries. It is fully distributed. A highlight of CABET is its non-uniform critical node sampling, called r r'-sampling, that selects landmarks to form boundary surfaces with bias toward nodes embodying salient topological features. Simulations show that CABET is able to extract a well-connected boundary in the presence of holes and shape variation, with performance superior to that of some state-of-the-art alternatives. In addition, we show how CABET benefits a range of sensor network applications including 3D skeleton extraction and 3D segmentation.

This paper presents a Connectivity-based and Anchor-free Three-dimensional Localization (CATL) scheme for large-scale sensor networks with concave regions. It distinguishes itself from previous work with a combination of three features: (1) it works for networks in both 2D and 3D spaces, possibly containing holes or concave regions; (2) it is anchor-free, and uses only connectivity information to faithfully recover the original network topology, up to scaling and rotation; (3) it does not depend on the knowledge of network boundaries, which suits it well to situations where boundaries are difficult to identify. The key idea of CATL is to discover the notch nodes, where shortest paths bend and hop-count-based distance starts to significantly deviate from the true Euclidean distance. An iterative protocol is developed that uses a em notch-avoiding multilateration mechanism to localize the network. Simulations show that CATL achieves accurate localization results with a moderate per-node message cost.