Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks316 from public data
- Crowd Counting and Density Estimation by Trellis Encoder-Decoder Networks366
This paper proposes a trellis encoder-decoder network (TEDnet) for crowd counting that achieves the best overall performance in terms of both density map quality and counting accuracy, and proposes a new combinatorial loss to enforce similarities in local coherence and spatial correlation between maps.
- 330
- 255
- 255
- Few-Shot Semantic Segmentation with Democratic Attention Networks241
This paper introduces the de-mocratized graph attention mechanism, which can activate more pixels on the object to establish a robust correspondence between support and query images, and proposes multi-scale guidance by designing a refinement fusion unit to fuse features from intermediate layers for the segmentation of the query image.
- Realizing the Metaverse with Edge Intelligence: A Match Made in Heaven238
The architecture of the Metaverse is discussed and mainly focus on motivating the convergence of edge intelligence and the infrastructure layer of theMetaverse, and major edge-based technological developments and their integration to support the metaverse engine are presented.
- Millimeter-Wave NOMA With User Grouping, Power Allocation and Hybrid Beamforming230
The proposed joint approach, including user grouping, hybrid beamforming and power allocation, outperforms the state-of-the-art schemes and the conventional mmWave orthogonal multiple access system in terms of achievable sum rate, and energy efficiency.
- 193
- 190
- Trajectory Design for UAV-Based Internet of Things Data Collection: A Deep Reinforcement Learning Approach185
An unmanned aerial vehicle (UAV)-assisted Internet of Things (IoT) system in a sophisticated 3-D environment, where the UAV’s trajectory is optimized to efficiently collect data from multiple IoT ground nodes, and presents a TD3-based trajectory design for completion time minimization (TD3-TDCTM) algorithm.
- Evolution of Chinese airport network182
It is found that although the topology of CAN has remained steady during the past few years, there are many dynamic switchings inside the network, which have changed the relative importance of airports and airlines.
- Millimeter-Wave Full-Duplex UAV Relay: Joint Positioning, Beamforming, and Power Control178
Simulation results confirm the superiority of the proposed positioning, beamforming, and power control method compared to three benchmark schemes and show that the proposed solution closely approaches a performance upper bound for mmWave FD-UAV systems.
- 163
- 153
- 151
- 147
- 147
- Path Planning for Single Unmanned Aerial Vehicle by Separately Evolving Waypoints144
The original objective and constraint functions of UAVs path planning are decomposed into a set of new evaluation functions, with which waypoints on a path can be evaluated separately and, thus, high-quality waypoints can be better exploited.
- Semantic Communication Meets Edge Intelligence142
This article introduces an edge-driven training, maintenance, and execution of semantic extraction (SE), and investigates how edge intelligence can be enhanced with SemCom through improving the generalization capabilities of intelligent agents at lower computation overheads and reducing the communication overhead of information exchange.
- 138
- Improving the network robustness against cascading failures by adding links136
Three link-adding strategies are compared and it is found that HBS is more effective than RLS to enhance the network robustness against cascades while the network exhibits the strongest robustness under LPS.
- 129
- 125
- 118
- Joint Power Control and Beamforming for Uplink Non-Orthogonal Multiple Access in 5G Millimeter-Wave Communications107
This paper investigates the combination of two key enabling technologies for the fifth generation wireless mobile communication, namely millimeter-wave (mm-wave) communications and non-orthogonal multiple access (NOMA), and proposes an analog beamforming problem under a constant-modulus constraint.
- NAS-Count: Counting-by-Density with Neural Architecture Search105
This work automates the design of counting models with Neural Architecture Search (NAS) and introduces an end-to-end searched encoder-decoder architecture, Automatic Multi-Scale Network (AMSNet), utilizing a counting-specific two-level search space.
- Projection Convolutional Neural Networks for 1-bit CNNs via Discrete Back Propagation100
This paper introduces projection convolutional neural networks with a discrete back propagation via projection (DBPP) to improve the performance of binarized neural networks (BNNs) and learns a set of diverse quantized kernels that compress the full-precision kernels in a more efficient way than those proposed previously.
- Cascading failure of interdependent networks with different coupling preference under targeted attack97
In this paper, the cascading failure model is extended to the case of interdependent networks, and the effect of coupling preference on systems robustness is intensively investigated.
- Analysis of the Chinese air route network as a complex network95
The Chinese air route network (CARN) is found to be a geographical network possessing exponential degree distribution, low clustering coefficient, large shortest path length and exponential spatial distance distribution that is obviously different from that of the Chinese airport network (CAN).
- Detection of Sudden Pedestrian Crossings for Driving Assistance Systems93
This paper proposes a three-level coarse-to-fine video-based framework that detects partially visible pedestrians just as they enter the camera view, with low false alarm rate and high speed and yields a performance better than that of state-of-the-art pedestrian detection while running at a frame rate of 55 fps.
- Vehicle Detection and Motion Analysis in Low-Altitude Airborne Video Under Urban Environment91
To speed up feature extraction and to retain additional global features in different scales for higher classification accuracy, a boosting light and pyramid sampling histogram of oriented gradients feature extraction method and a spatio-temporal appearance-related similarity measure are proposed.
- 89
- Cooperative pursuit of unauthorized UAVs in urban airspace via Multi-agent reinforcement learning80
This paper model cooperative pursuit as a pursuit-evasion game problem (PEG) and proposes a multi-agent reinforcement learning (MARL) based approach to solve the problem efficiently, which incorporates novel cellular-enabled parameter sharing and curriculum learning schemes to enhance the capability of pursuer UAVs in capturing faster unauthorized Uavs in urban airspace.
- 77
- Output Constraint Transfer for Kernelized Correlation Filter in Tracking77
Extensive experiments show that the proposed output constraint transfer (OCT) method significantly improves KCF, and achieves better performance than other state-of-the-art trackers.
- 75
- Integrating neighborhoods in the evaluation of fitness promotes cooperation in the spatial prisoner’s dilemma game74
It is found that considering the environment, i.e., integrating neighborhoods in the evaluation of fitness, promotes cooperation, and the definition of individual fitness is reconsidered which integrates the environment with the traditional individual payoffs.
- Image Registration and Fusion of Visible and Infrared Integrated Camera for Medium-Altitude Unmanned Aerial Vehicle Remote Sensing72
A new image fusion algorithm is designed based on a pulse coupled neural network and nonsubsampled contourlet transform to meet the special requirements of preserving color information, adding infrared brightness information, improving spatial resolution, and highlighting target areas for unmanned aerial vehicle (UAV) applications.
- 68
- 67
- 67
- Heterogeneous Strategy Particle Swarm Optimization65
This brief presents a heterogeneous strategy PSO (HSPSO), in which a proportion of particles adopts a fully informed strategy to enhance the converging speed while the rest is singly informed to maintain the diversity.
- The networked evolutionary algorithm: A network science perspective63
This work systematically explore the evolutionary algorithm as a networked interaction system, where nodes represent information process units and connections denote information transmission links, and implements it in three typical evolutionary algorithms, namely in the genetic algorithm, the particle swarm optimization algorithm, and in the differential evolution algorithm.
- Identifying vital edges in Chinese air route network via memetic algorithm63
This work systematically explores the robustness of the Chinese air route network, and identifies the vital edges which form the backbone of Chinese air transportation system.
- 61
- Analysis of the Chinese railway system as a complex network60
This paper encapsulates CRN into a multi-layer infrastructure with a core-periphery structure via the ‘‘weighted k-core decomposition” method, and finds the reasonability of cities contained in the core layer is generally consistent with the hub set identified by the gravity model.
- Pheromone routing protocol on a scale-free network60
A routing strategy for network systems based on the local information of "pheromone" that performs much better than the local routing strategy, which is demonstrated by a larger value of the critical generating rate.
- Transfer learning for pedestrian detection59
Compared to the traditional pedestrian detection methods, the proposed algorithm can adapt to different scenes for detecting pedestrians and can obtain better performance as compared to other previous methods.
- Computationally efficient attack design for robustness analysis of air transportation networks58
A new exploration/exploitation search technique for a computationally efficient attacking model, adapted from general game playing, that contributes to scalable analysis of robustness, not only for air transportation, but also for networks in general.
- One-two-one networks for compression artifacts reduction in remote sensing57
An in-depth investigation into the OTO architecture based on the Taylor expansion is provided, which shows that these two kinds of information can be fused in a nonlinear scheme to gain more capacity of handling complicated image compression artifacts, especially the blocking effect in compression.
- 54
- 54
- 54
- 53
- Evolution of cooperation among mobile agents with heterogenous view radii53
It is found that the cooperation level is remarkably promoted when the heterogeneity of view radii is considered, and the degree distribution of the system is investigated to explain this interesting phenomenon.
- 52
- Structural properties of the Chinese air transportation multilayer network51
The results show that the high clustering coefficient, short characteristic path length and large collection of reachable destinations of the Chinese ATMN can only emerge when several layers are merged together.
- 51
- Quantum-Secured Space-Air-Ground Integrated Networks: Concept, Framework, and Case Study50
This article proposes quantum-secured SAGIN (Q-SAGIN) which are feasible to achieve proven secure communications using quantum mechanics to protect data channels between space, aerial, and ground nodes and proposes a universal QKD service provisioning framework to minimize the cost ofQKD services under the uncertainty and dynamics of communications in Q- SAGIN.
- 50
- Selecting Key Poses on Manifold for Pairwise Action Recognition50
A novel approach for key poses selection is proposed, which models the descriptor space utilizing a manifold learning technique to recover the geometric structure of the descriptors on a lower dimensional manifold and develops a PageRank-based centrality measure.
- Vehicle detection and tracking in airborne videos by multi-motion layer analysis50
A new framework of multi-motion layer analysis is proposed to detect and track moving vehicles in airborne platform that is more efficient and more robust to partial occlusion, and able to meet the demand of real time application due to its computational simplicity.
- 49
- 49
- Traffic dynamics on coupled spatial networks47
This paper extensively investigate the traffic process on these coupled spatial networks and focuses on the effect of the parameter α, the speed ratio between two networks, indicating that cooperation between the coupled networks will induce the highest network capacity at an optimal α .
- Effect of Adaptive Delivery Capacity on Networked Traffic Dynamics47
It is found that the network capacity monotonously increases with the increment of φ, but there exists an optimal value of parameter φ leading to the highest transportation efficiency measured by average travelling time (〈T〉).
- 46
- Pylon line spatial correlation assisted transmission line detection45
Compared with three other line-based TLD methods, the experimental results demonstrate that the proposed method can obtain better detection performance with higher detection rates and much lower false alarm rates.
- 44
- Survey on Near-Space Information Networks: Channel Modeling, Transmission, and Networking Perspectives43
The characteristics of NSINs are highlighted and the promising use cases of NSINs are presented and the latest advances in NSINs in the research areas of channel modeling, transmission, and networking are discussed.
- 43
- 43
- On Node Criticality in Air Transportation Networks43
The evaluation on the international air transportation country network suggests that some countries, e.g., United States, France, and Germany, are critical from all three perspectives, while other countries have a very specific influence, by controlling the passenger traffic of their neighborhood countries.
- A knowledge-transfer-based learning framework for airspace operation complexity evaluation42
A new sector operation complexity evaluation framework based on knowledge transfer specifically for small-training-sample environment is proposed, able to effectively mine knowledge hidden within the samples of the target sector and properly handle the integration between the knowledge derived from different sectors.
- Random sketch learning for deep neural networks in edge computing40
A universal compressing-while-training framework that directly learns a compact model and, most importantly, enables computationally efficient on-device learning for tiny artificial intelligence in resource-constrained devices is built.
- QoE-Driven Video Transmission: Energy-Efficient Multi-UAV Network Optimization39
A novel network optimization algorithm is proposed to solve this challenging problem, in which a Lyapunov technique is first explored to decompose the sequential-decision problem into several repeatedly optimized sub-problems to avoid the curse of dimensionality.
- 38
- 38
- 38
- Promotion of cooperation in aspiration-based spatial prisoner’s dilemma game38
It is found that the cooperator frequency is not sensitive to the initial condition but the evolutionary patterns can be quite different and the evolution time series show the so-called "ping-pong effect" for both high and low aspiration levels.
- 38
- 37
- 37
- 37
- 37
- Fast Randomized-MUSIC for Mm-Wave Massive MIMO Radars36
This work develops a fast randomized-MUSIC (R-M USIC) algorithm, which exploits the random matrix sketching to estimate the signal subspace via approximated computation, and resolves the long-standing contradiction in complexity and accuracy of MIMO radar signal processing.
- ATCEM: a synthetic model for evaluating air traffic complexity36
An air traffic complexity evaluation model with integrated classification using computational intelligence (ATCEM) that outperforms a number of state-of-the-art models and could be effectively estimated.
- Associated evolution of a support vector machine-based classifier for pedestrian detection36
This paper proposes an evolutionary method to simultaneously optimize the feature set and the parameters for the SVM classifier and shows that the proposed approach leads to higher detection accuracy and shorter detection time.
- Feeling of Presence Maximization: mmWave-Enabled Virtual Reality Meets Deep Reinforcement Learning34
A parallel echo state network (ESN) learning method to predict users’ tracking information by training fresh and historical tracking samples separately collected by APs is developed and a deep reinforcement learning (DRL) based optimization algorithm is proposed to solve the formulated problem.
- How Should I Orchestrate Resources of My Slices for Bursty URLLC Service Provision?34
This paper proposes to orchestrate network resources for a slicing system to guarantee more reliable bursty URLLC transmission and develops a resource optimization algorithm based on a sample average approximate technique and a distributed optimization method with provable performance guarantees.
- 34
- 34
- Ego motion guided particle filter for vehicle tracking in airborne videos34
A new framework based on Kanade-Lucas-Tomasi features and particle filter is proposed and it is demonstrated that the proposed method outperformed the other tracking methods.
- 33
- RAN Slicing for Massive IoT and Bursty URLLC Service Multiplexing: Analysis and Optimization33
A slice resource optimization (SRO) algorithm exploiting relaxation and approximation with provable tightness and error bound is proposed to mitigate the optimization problem of the RAN slicing for mIoT and bursty ultra-reliable and low latency communications (URLLC) service multiplexing.
- 32
- Analysis of flight conflicts in the Chinese air route network32
It is shown that air conflicts are distributed heterogeneously along the waypoints of the Chinese air route network, and the frequency of flight conflicts follows an exponential distribution.
- Using computational intelligence for large scale air route networks design32
MAPP has been implemented as a module in the software that is currently used for ARN planning in China and outperformed an existing approach to CWLP as well as three well-known Multi-Objective Evolutionary Algorithms (MOEAs).
- 31
- Measuring air traffic complexity based on small samples31
An ensemble learning model is proposed for measuring air traffic complexity within a sector based on small samples and demonstrates the advantages of this model for small-sample-based traffic complexity evaluation over other state-of-the-art methods.
- Vision-Aided RAIM: A New Method for GPS Integrity Monitoring in Approach and Landing Phase31
A new RAIM method, named vision-aided RAIM (VA-RAIM), is proposed for GPS integrity monitoring in the approach and landing phase, which outperforms the conventional RAIM with a higher level of availability and fault detection rate.
- Object-aware power line detection using color and near-infrared images31
A new definition of power line is first proposed to present its object-aware properties; and then a cascaded PLD scheme is devised with line detection, region filtration, and object validation based on corresponding image cues in joint color (RGB) and near-infrared (NIR) images.
- 30
- Long-Short-Term Features for Dynamic Scene Classification30
A unified framework to extract spatial and temporal features for dynamic scene representation is proposed and two variants of deep convolutional neural networks are deployed to encode spatial appearance and short- term dynamics into short-term deep features (STDF).
- 29
- 29
- 28
- 28
- 28
- Ultra-Fast Accurate AoA Estimation via Automotive Massive-MIMO Radar28
This work proposes two efficient methods, to accomplish fast subspace computation and accurate angle of arrival (AoA) acquisition, and establishes the theoretical bounds of the proposed methods, which ensure the accuracy of the approximated pseudo-spectrum.
- 28
- 28
- 28
- Evolutionary Games on Weighted Newman–Watts Small-World Networks28
This work investigates the evolutionary prisoner's dilemma game (PDG) on weighted Newman–Watts (NW) networks and employs the Gini coefficient to study the wealth distribution in the population.
- DNN Task Assignment in UAV Networks: A Generative AI Enhanced Multiagent Reinforcement Learning Approach27
This work proposes a two-stage optimization method for flight path planning and task allocation based on a mother-child UAV swarm system and introduces a novel DNN task assignment algorithm that combines multiagent deep deterministic policy gradient (MADDPG) and generative diffusion models (GDMs), named GDM-MADDPG.
- Attentional Alignment Networks.27
This work proposes attentional alignment networks (AAN), a novel end-to-end convolutional architecture for direct face alignment without relying on cascaded regression that consistently delivers high performance on all datasets, surpassing previous methods by large margins.
- 27
- Multiple Auxiliaries Assisted Airborne Power Line Detection26
Experimental results show that the proposed method surpasses the state-of-the-art power line detection methods, both in terms of detection accuracy and false alarm probability.
- The effect of attack cost on network robustness26
In this paper, the factor of attack cost associated with network robustness is taken into account and it is found that the performance of different attack strategies is sensitive to the total attack cost.
- 26
- 25
- Enhancing Routing Performance Through Trajectory Planning With DRL in UAV-Aided VANETs24
A heterogeneous UAV-aided VANET framework that leverages uncrewed aerial vehicles (UAVs), also known as autonomous aerial vehicles, to enhance data transmission is introduced, demonstrating the potential to enhance VANET performance, while also providing valuable insights for the development of intelligent transportation systems and related fields.
- 24
- 24
- 24
- Hyperspectral imaging from a raw mosaic image with end-to-end learning24
An end-to-end deep learning method to reconstruct hyperspectral images directly from a raw mosaic image, which saves the separate demosaicing process required by other methods, which reconstructs the full-resolution RGB data from the raw mosaicImage, reduces computational complexity and accumulative error.
- An evolutionary approach for dynamic single-runway arrival sequencing and scheduling problem24
This paper proposes an evolutionary approach to aircraft arrival sequencing and scheduling problem, namely dynamic sequence searching and evaluation, which employs an estimation of distribution algorithm and a heuristic search method to seek the optimal landing sequence of flights.
- Multicast eMBB and Bursty URLLC Service Multiplexing in a CoMP-Enabled RAN23
An iterative algorithm with provable performance guarantees is developed to obtain solutions to single timescale problems and aggregate obtained solutions into those of the multi-timescale problem.
- Characterizing Flight Delay Profiles with a Tensor Factorization Framework23
A probabilistic framework for high-dimensional historical flight data is proposed using a latent class model and the effectiveness of this framework is demonstrated using air traffic data from 224 airports in China during 2014–2017.
- Multi-source off-grid DOA estimation with single snapshot using non-uniform linear arrays23
This work has derived the analytical expression of the mean square error (MSE) of the proposed OGDOA estimation method and verified that the MSE approaches the Cram e ´ r-Rao lower Bound (CRB) when the signal-to-noise ratio (SNR) increases.
- 23
- Travel time analysis in the Chinese coupled aviation and high-speed rail network23
The results indicate that sufficient frequencies of flights/HSR trains have led to high multi-modal accessibility across different time periods of the day, and that highly intermodal air-HSR mobility pathways can be extremely important to link small cities to urban mega-region hubs.
- Propagation Index on Airport Delays22
A new measure, the propagation index, is proposed, to effectively analyze the interrelationship among airports in relation to flight delays and is highly correlated with observed airport delays.
- 22
- Fresh, Fair and Energy-Efficient Content Provision in a Private and Cache-Enabled UAV Network21
A novel algorithm based on the techniques of subproblem decomposition and convex approximation is proposed to tackle the challenging mixed-integer non-convex subproblems of a private and cache-enabled unmanned aerial vehicle (UAV) network for content provision.
- Power Control for a URLLC-Enabled UAV System Incorporated With DNN-Based Channel Estimation21
A novel power control algorithm is proposed, which builds analytically tractable channel models based on DNN estimation results and explores semidefinite relaxation (SDR) with provable performance guarantees to tackle the non-convexity of the UAV system.
- 21
- 20
- 20
- Sequence searching and evaluation: a unified approach for aircraft arrival sequencing and scheduling problems20
An approach that is applicable to a number of different problem formulations of ASS is proposed, and a Sequence Searching and Evaluation (SSE) approach is developed for the constrained permutation-based problem.
- Networking of Internet of UAVs: Challenges and Intelligent Approaches19
It is argued that I-UAV networking can be classified into two categories: quality-of-service (QoS) driven networking and quality-of-experience (QoE) driven networking.
- 19
- 18
- 18
- 18
- Optimal resource allocation in interdependent networks18
A novel particle swarm optimization algorithm with an information feedback mechanism to achieve better performance on complex problems and produces a robust resource allocation pattern for interdependent networks, where middle-degree nodes are assigned more tolerance.
- 18
- Visual Attention Accelerated Vehicle Detection in Low-Altitude Airborne Video of Urban Environment18
A new and efficient method composed by two stages, attention focus extraction and vehicle classification is proposed, which can obtain better performance in terms of higher detection rate and lower false positive rate, while meeting the needs of real-time application.
- 17
- A semi-supervised zero-shot image classification method based on soft-target17
A novel Soft-Target Semi-supervised Classification (STSC) model with an autoencoder network is leveraged, where both labeled seen data from the seen categories and unlabeled ancillary data collected from Internet or other datasets are employed as two branches, respectively.
- Attentive Temporal Pyramid Network for Dynamic Scene Classification17
The proposed attentive temporal pyramid network (ATP-Net) leverages the strength of attention mechanism to select the most relevant frame features and the ability of kernels to achieve optimal feature fusion for discriminative representations of dynamic scenes.
- Hierarchical incorporation of shape and shape dynamics for flying bird detection17
A novel hierarchical feature model exploiting shape and shape dynamics to improve the ability of representing a flying bird, and then applies it to the FBD problem to show that the proposed method is effective and outperforms three other existing vision-based FBD approaches.
- 17
- Network Topology Inference Based on Timing Meta-Data16
A new Expectation Maximization (EM)-based algorithm is introduced– EM-causality discovery algorithm (EM-CDA)– which treats packet losses as latent variables and iterates between the estimation of packet losses and the evaluation of causality metrics.
- 16
- Alignment Enhancement Network for Fine-grained Visual Categorization16
A novel Alignment Enhancement Network (AENet) is proposed, including two-level alignments, Cross-layer Alignment (CLA) and Cross-image Aligned (CIA), which exploits the cross-layer relationship between low-level spatial information and high-level semantic information, which contributes to cross- layer feature aggregation to improve the capacity of feature representation for input images.
- An Improved Faster R-CNN for UAV-Based Catenary Support Device Inspection16
Catenary support device inspection is of crucial importance for ensuring safety and reliability of railway systems and visual detection tasks of catenary support devices defect are performed at present.
- 16
- 16
- Pilot-Aided Joint Time Synchronization and Channel Estimation for OTFS15
This letter proposes a pilot-aided joint time synchronization and channel estimation algorithm for orthogonal time frequency space (OTFS) systems that employs a maximum length sequence rather than an isolated signal as the pilot.
- 15
- 14
- Object detection and tracking under Complex environment using deep learning‐based LPM14
A novel local probability model (LPM)-based mean shift (MS) algorithm is proposed to integrate object detection and tracking and demonstrates that the proposed method is superior to the colour histograms based MS and histograms of oriented gradients based MS in terms of robustness and tracking accuracy.
- 14
- Region tracking control for high‐order multi‐agent systems in restricted space14
It is proved that under the adaptive NN control, the tracking error of each agent can converge to an adjustable neighbourhood of the origin, although some of them do not access the desired region directly.
- Tracking vehicles as groups in airborne videos14
A new collaborative tracking framework to track vehicles as groups takes into account the relevance between vehicles and reduces the impact of camera vibration, and demonstrates that the proposed method has better performance in terms of tracking speed and tracking accuracy.
- 14
- 14
- 14
- 13
- 13
- 13
- 13
- 13
- The effect of packet lifetime on scale-free network information traffic13
This work model the information traffic on scale-free networks considering the packets’ limited lifetime and finds that the network systems undergo a continuous transition from stable state to explored state at a critical point Rd.
- The effect of TTL on the information traffic13
This paper introduces TTL into the information traffic model on Barabasi–Albert scale-free networks under local routing strategy and focuses on its effect on the network capacity measured by the critical point (Rc) of phase transition from free flow to congestion.
- 12
- Leveraging Spatial Context Disparity for Power Line Detection12
The proposed power line detection method, which can automatically choose the optimal auxiliaries, is effective and has the potential for practical use in ensuring the flight safety of unmanned air vehicles (UAVs) in the low flying environment.
- 12
- 11
- 11
- Model-Free Tracking With Deep Appearance and Motion Features Integration11
A real-time model-free object tracking approach is designed in this work relying on Convolutional Neural Networks (CNNs), which are deeply integrated by the proposed AMNet, which is an end-to-end offline trained two-stream network.
- 11
- THE EFFECT OF PINNING CONTROL ON EVOLUTIONARY PRISONER'S DILEMMA GAME11
D degree-basedPinning control and random pinning control are introduced into the evolutionary prisoner's dilemma game on scale-free networks, and the effects of control mechanism and control cost on the evolution are studied.
- Energy-Efficient URLLC Service Provision via a Near-Space Information Network10
This paper investigates the problem of deploying a RIS-integrated NSIN to provide energy-efficient, ultra-reliable and low-latency communications (URLLC) services and proposes a joint resource allocation algorithm to handle these challenges.
- 10
- 10
- P4Q: Learning to Prompt for Quantization in Visual-language Models9
This work proposes a method that balances fine-tuning and quantization named ``Prompt for Quantization'' (P4Q), in which a lightweight architecture is designed to leverage contrastive loss supervision to enhance the recognition performance of a PTQ model.
- 9
- 9
- 9
- 9
- 9
- 8
- 8
- Representation Disparity-aware Distillation for 3D Object Detection8
A novel representation disparity-aware distillation (RDD) method is presented to address the representation disparity issue and reduce performance gap between compact students and over-parameterized teachers by building the RDD from an innovative perspective of information bottleneck (IB).
- 8
- 8
- 8
- Multi-scale Supervised Attentive Encoder-Decoder Network for Crowd Counting8
A new KL-divergence-based distribution loss is designed to supervise the scale-aware structural differences between two density maps, which complements the pixel-isolated MSE loss and better optimizes AEDN to generate high-quality density maps.
- 8
- 8
- 8
- 8
- 8
- 8
- 7
- 7
- 7
- 7
- 7
- 6
- 6
- 6
- 6
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- 5
- 4
- 4
- 4
- Deep Collaborative Tracking Networks.4
The DCTN is the first generalized framework to model motion and appearance information with deep learning for object tracking and can achieve high tracking performance, which demonstrates the great effectiveness of exploring both motion and appearance information for visual object tracking.
- 4
- 4
- 4
- 4
- Noise-Robust tiny object localization with flows3
This work proposes Tiny Object Localization with Flows (TOLF), a noise-robust localization framework leveraging normalizing flows for flexible error modeling and uncertainty-guided optimization, enabling robust learning under noisy supervision.
- 3
- 3
- 3
- 3
- Heterogeneous Generative Knowledge Distillation with Masked Image Modeling3
This paper develops the first Heterogeneous Generative Knowledge Distillation (H-GKD) based on MIM, which can efficiently transfer knowledge from large Transformer models to small CNN-based models in a generative self-supervised fashion.
- DCP–NAS: Discrepant Child–Parent Neural Architecture Search for 1-bit CNNs3
This paper introduces Discrepant Child–Parent Neural Architecture Search (DCP–NAS), a new framework of searching the 1-bit model under the supervision of a real-valued model (Parent), and proposes a decoupled optimization method to search an optimized architecture.
- 3
- 3
- 3
- 3
- 3
- 3
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- ASiam: adaptive Siamese regression tracking with adversarial template generation and motion‐based failure recovery2
A new adaptive Siamese (ASiam) tracker is proposed, which integrates a novel adversarial template generation module and a motion-based failure recovery module and is optimised via adversarial learning to achieve accurate appearance prediction and sharp template quality.
- 2
- Special issue on image feature detection and description2
The aim of this special issue is to bring together the state-of-the research results on image feature detection and description and its applications and to accept about 30-40 papers.
- 2
- 2
- 2
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- Big2Small: Learning from masked image modelling with heterogeneous self‐supervised knowledge distillation1
The authors develop the first heterogeneous self‐supervised knowledge distillation (HSKD) based on MIM, which can efficiently transfer knowledge from large transformer models to small CNN‐based models in a self‐supervised fashion.
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- –
- –
- P4Q: Learning to Prompt for Quantization in Low-Bit CLIP–
The “Prompt for Quantization” (P4Q) is proposed, by integrating PTQ with Parameter-Efficient Fine-Tuning (PEFT) techniques, and demonstrates that P4Q significantly enhances the performance of low-bit CLIP while reducing deployment costs.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-16. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.