arXiv subjects

Long Hu

Publications and source records attributed to Long Hu.

At least 37 records · Page 2Linked to original sources

arXiv2024

Multimodal Physiological Signals Representation Learning via Multiscale Contrasting for Depression Recognition

Depression recognition based on physiological signals such as functional near-infrared spectroscopy (fNIRS) and electroencephalogram (EEG) has made considerable progress. However, most existing studies ignore the complementarity and semantic consistency of multimodal physiological signals under the same stimulation task in complex spatio-temporal patterns. In this paper, we introduce a multimodal physiological signals representation learning framework using Siamese architecture via multiscale contrasting for depression recognition (MRLMC). First, fNIRS and EEG are transformed into different but correlated data based on a time-domain data augmentation strategy. Then, we design a spatio-temporal contrasting module to learn the representation of fNIRS and EEG through weight-sharing multiscale spatio-temporal convolution. Furthermore, to enhance the learning of semantic representation associated with stimulation tasks, a semantic consistency contrast module is proposed, aiming to maximize the semantic similarity of fNIRS and EEG. Extensive experiments on publicly available and self-collected multimodal physiological signals datasets indicate that MRLMC outperforms the state-of-the-art models. Moreover, our proposed framework is capable of transferring to multimodal time series downstream tasks.

cs.LG↗

arXiv2024

Bilinear-Convolutional Neural Network Using a Matrix Similarity-based Joint Loss Function for Skin Disease Classification

In this study, we proposed a model for skin disease classification using a Bilinear Convolutional Neural Network (BCNN) with a Constrained Triplet Network (CTN). BCNN can capture rich spatial interactions between features in image data. This computes the outer product of feature vectors from two different CNNs by a bilinear pooling. The resulting features encode second-order statistics, enabling the network to capture more complex relationships between different channels and spatial locations. The CTN employs the Triplet Loss Function (TLF) by using a new loss layer that is added at the end of the architecture called the Constrained Triplet Loss (CTL) layer. This is done to obtain two significant learning objectives: inter-class categorization and intra-class concentration with their deep features as often as possible, which can be effective for skin disease classification. The proposed model is trained to extract the intra-class features from a deep network and accordingly increases the distance between these features, improving the model's performance. The model achieved a mean accuracy of 93.72%.

cs.CV↗

arXiv2024

MiniGPT-3D: Efficiently Aligning 3D Point Clouds with Large Language Models using 2D Priors

Large 2D vision-language models (2D-LLMs) have gained significant attention by bridging Large Language Models (LLMs) with images using a simple projector. Inspired by their success, large 3D point cloud-language models (3D-LLMs) also integrate point clouds into LLMs. However, directly aligning point clouds with LLM requires expensive training costs, typically in hundreds of GPU-hours on A100, which hinders the development of 3D-LLMs. In this paper, we introduce MiniGPT-3D, an efficient and powerful 3D-LLM that achieves multiple SOTA results while training for only 27 hours on one RTX 3090. Specifically, we propose to align 3D point clouds with LLMs using 2D priors from 2D-LLMs, which can leverage the similarity between 2D and 3D visual information. We introduce a novel four-stage training strategy for modality alignment in a cascaded way, and a mixture of query experts module to adaptively aggregate features with high efficiency. Moreover, we utilize parameter-efficient fine-tuning methods LoRA and Norm fine-tuning, resulting in only 47.8M learnable parameters, which is up to 260x fewer than existing methods. Extensive experiments show that MiniGPT-3D achieves SOTA on 3D object classification and captioning tasks, with significantly cheaper training costs. Notably, MiniGPT-3D gains an 8.12 increase on GPT-4 evaluation score for the challenging object captioning task compared to ShapeLLM-13B, while the latter costs 160 total GPU-hours on 8 A800. We are the first to explore the efficient 3D-LLM, offering new insights to the community. Code and weights are available at https://github.com/TangYuan96/MiniGPT-3D.

cs.CV↗

arXiv2023

TriGait: Aligning and Fusing Skeleton and Silhouette Gait Data via a Tri-Branch Network

Gait recognition is a promising biometric technology for identification due to its non-invasiveness and long-distance. However, external variations such as clothing changes and viewpoint differences pose significant challenges to gait recognition. Silhouette-based methods preserve body shape but neglect internal structure information, while skeleton-based methods preserve structure information but omit appearance. To fully exploit the complementary nature of the two modalities, a novel triple branch gait recognition framework, TriGait, is proposed in this paper. It effectively integrates features from the skeleton and silhouette data in a hybrid fusion manner, including a two-stream network to extract static and motion features from appearance, a simple yet effective module named JSA-TC to capture dependencies between all joints, and a third branch for cross-modal learning by aligning and fusing low-level features of two modalities. Experimental results demonstrate the superiority and effectiveness of TriGait for gait recognition. The proposed method achieves a mean rank-1 accuracy of 96.0% over all conditions on CASIA-B dataset and 94.3% accuracy for CL, significantly outperforming all the state-of-the-art methods. The source code will be available at https://github.com/feng-xueling/TriGait/.

cs.CV↗

arXiv2023

Explore More Guidance: A Task-aware Instruction Network for Sign Language Translation Enhanced with Data Augmentation

Sign language recognition and translation first uses a recognition module to generate glosses from sign language videos and then employs a translation module to translate glosses into spoken sentences. Most existing works focus on the recognition step, while paying less attention to sign language translation. In this work, we propose a task-aware instruction network, namely TIN-SLT, for sign language translation, by introducing the instruction module and the learning-based feature fuse strategy into a Transformer network. In this way, the pre-trained model's language ability can be well explored and utilized to further boost the translation performance. Moreover, by exploring the representation space of sign language glosses and target spoken language, we propose a multi-level data augmentation scheme to adjust the data distribution of the training set. We conduct extensive experiments on two challenging benchmark datasets, PHOENIX-2014-T and ASLG-PC12, on which our method outperforms former best solutions by 1.65 and 1.42 in terms of BLEU-4. Our code is published at https://github.com/yongcaoplus/TIN-SLT.

cs.CL↗

arXiv2022

CasFusionNet: A Cascaded Network for Point Cloud Semantic Scene Completion by Dense Feature Fusion

Semantic scene completion (SSC) aims to complete a partial 3D scene and predict its semantics simultaneously. Most existing works adopt the voxel representations, thus suffering from the growth of memory and computation cost as the voxel resolution increases. Though a few works attempt to solve SSC from the perspective of 3D point clouds, they have not fully exploited the correlation and complementarity between the two tasks of scene completion and semantic segmentation. In our work, we present CasFusionNet, a novel cascaded network for point cloud semantic scene completion by dense feature fusion. Specifically, we design (i) a global completion module (GCM) to produce an upsampled and completed but coarse point set, (ii) a semantic segmentation module (SSM) to predict the per-point semantic labels of the completed points generated by GCM, and (iii) a local refinement module (LRM) to further refine the coarse completed points and the associated labels from a local perspective. We organize the above three modules via dense feature fusion in each level, and cascade a total of four levels, where we also employ feature fusion between each level for sufficient information usage. Both quantitative and qualitative results on our compiled two point-based datasets validate the effectiveness and superiority of our CasFusionNet compared to state-of-the-art methods in terms of both scene completion and semantic segmentation. The codes and datasets are available at: https://github.com/JinfengX/CasFusionNet.

cs.CV↗

arXiv2021

Equivalent one-dimensional first-order linear hyperbolic systems and range of the minimal null control time with respect to the internal coupling matrix

In this paper, we are interested in the minimal null control time of one-dimensional first-order linear hyperbolic systems by one-sided boundary controls. Our main result is an explicit characterization of the smallest and largest values that this minimal null control time can take with respect to the internal coupling matrix. In particular, we obtain a complete description of the situations where the minimal null control time is invariant with respect to all the possible choices of internal coupling matrices. The proof relies on the notion of equivalent systems, in particular the backstepping method, a canonical $LU$-decomposition for boundary coupling matrices and a compactness-uniqueness method adapted to the null controllability property.

math.OC↗

arXiv2020

Null controllability and finite-time stabilization in minimal time of one-dimensional first-order $2 \times 2$ linear hyperbolic systems

The goal of this article is to present the minimal time needed for the null controllability and finite-time stabilization of one-dimensional first-order $2 \times 2$ linear hyperbolic systems. The main technical point is to show that we cannot obtain a better time. The proof combines the backstepping method with the Titchmarsh convolution theorem.

math.OC↗

arXiv2020

Boundary stabilization in finite time of one-dimensional linear hyperbolic balance laws with coefficients depending on time and space

In this article we are interested in the boundary stabilization in finite time of one-dimensional linear hyperbolic balance laws with coefficients depending on time and space. We extend the so called "backstepping method" by introducing appropriate time-dependent integral transformations in order to map our initial system to a new one which has desired stability properties. The kernels of the integral transformations involved are solutions to non standard multi-dimensional hyperbolic PDEs, where the time dependence introduces several new difficulties in the treatment of their well-posedness. This work generalizes previous results of the literature, where only time-independent systems were considered.

math.OC↗

arXiv2019

Femtosecond relativistic electron beam with reduced timing jitter from THz-driven beam compression

We propose and demonstrate a novel method to reduce the pulse width and timing jitter of a relativistic electron beam through THz-driven beam compression. In this method the longitudinal phase space of a relativistic electron beam is manipulated by a linearly polarized THz pulse in a dielectric tube such that the bunch tail has a higher velocity than the bunch head, which allows simultaneous reduction of both pulse width and timing jitter after passing through a drift. In this experiment, the beam is compressed by more than a factor of four from 130 fs to 28 fs with the arrival time jitter also reduced from 97 fs to 36 fs, opening up new opportunities in using pulsed electron beams for studies of ultrafast dynamics. This technique extends the well known rf buncher to the THz frequency and may have a strong impact in accelerator and ultrafast science facilities that require femtosecond electron beams with tight synchronization to external lasers.

physics.acc-ph↗

arXiv2019

A Sustainable Multi-modal Multi-layer Emotion-aware Service at the Edge

Limited by the computational capabilities and battery energy of terminal devices and network bandwidth, emotion recognition tasks fail to achieve good interactive experience for users. The intolerable latency for users also seriously restricts the popularization of emotion recognition applications in the edge environments such as fatigue detection in auto-driving. The development of edge computing provides a more sustainable solution for this problem. Based on edge computing, this article proposes a multi-modal multi-layer emotion-aware service (MULTI-EASE) architecture that considers user's facial expression and voice as a multi-modal data source of emotion recognition, and employs the intelligent terminal, edge server and cloud as multi-layer execution environment. By analyzing the average delay of each task and the average energy consumption at the mobile device, we formulate a delay-constrained energy minimization problem and perform a task scheduling policy between multiple layers to reduce the end-to-end delay and energy consumption by using an edge-based approach, further to improve the users' emotion interactive experience and achieve energy saving in edge computing. Finally, a prototype system is also implemented to validate the architecture of MULTI-EASE, the experimental results show that MULTI-EASE is a sustainable and efficient platform for emotion analysis applications, and also provide a valuable reference for dynamic task scheduling under MULTI-EASE architecture.

cs.NI↗

arXiv2019

AI-Skin : Skin Disease Recognition based on Self-learning and Wide Data Collection through a Closed Loop Framework

There are a lot of hidden dangers in the change of human skin conditions, such as the sunburn caused by long-time exposure to ultraviolet radiation, which not only has aesthetic impact causing psychological depression and lack of self-confidence, but also may even be life-threatening due to skin canceration. Current skin disease researches adopt the auto-classification system for improving the accuracy rate of skin disease classification. However, the excessive dependence on the image sample database is unable to provide individualized diagnosis service for different population groups. To overcome this problem, a medical AI framework based on data width evolution and self-learning is put forward in this paper to provide skin disease medical service meeting the requirement of real time, extendibility and individualization. First, the wide collection of data in the close-loop information flow of user and remote medical data center is discussed. Next, a data set filter algorithm based on information entropy is given, to lighten the load of edge node and meanwhile improve the learning ability of remote cloud analysis model. In addition, the framework provides an external algorithm load module, which can be compatible with the application requirements according to the model selected. Three kinds of deep learning model, i.e. LeNet-5, AlexNet and VGG16, are loaded and compared, which have verified the universality of the algorithm load module. The experiment platform for the proposed real-time, individualized and extensible skin disease recognition system is built. And the system's computation and communication delay under the interaction scenario between tester and remote data center are analyzed. It is demonstrated that the system we put forward is reliable and effective.

cs.CV↗

arXiv2019

Ready Player One: UAV Clustering based Multi-Task Offloading for Vehicular VR/AR Gaming

With rapid development of unmanned aerial vehicle (UAV) technology, application of the UAVs for task offloading has received increasing interest in the academia. However, real-time interaction between one UAV and the mobile edge computing (MEC) node is required for processing the tasks of mobile end users, which significantly increases the system overhead and is unable to meet the demands of large-scale artificial intelligence (AI) based applications. To tackle this problem, in this article, we propose a new architecture for UAV clustering to enable efficient multi-modal multi-task task offloading. By the proposed architecture, the computing, caching and communication resources are collaboratively optimized using AI based decision-making. This not only increases the efficiency of UAV clusters, but also provides insight into the fusion of computation and communication.

cs.NI↗

arXiv2019

Minimal time for the exact controllability of one-dimensional first-order linear hyperbolic systems by one-sided boundary controls

In this article we study the minimal time for the exact controllability of one-dimensional first-order linear hyperbolic systems when all the controls are acting on the same side of the boundary. We establish an explicit and easy-to-compute formula for this time with respect to all the coupling parameters of the system. The proof relies on the introduction of a canonical $UL$-decomposition and the compactness-uniqueness method.

math.OC↗

arXiv2019

Terahertz oscilloscope for recording time information of ultrashort electron beams

We propose and demonstrate a Terahertz (THz) oscilloscope for recording time information of an ultrashort electron beam. By injecting a laser-driven THz pulse with circular polarization into a dielectric tube, the electron beam is swept helically such that the time information is uniformly encoded into the angular distribution that allows one to characterize both the temporal profile and timing jitter of an electron beam. The dynamic range of the measurement in such a configuration is significantly increased compared to deflection with a linearly polarized THz pulse. With this THz oscilloscope, nearly 50-fold longitudinal compression of a relativistic electron beam to about 15 fs (rms) is directly visualized with its arrival time determined with 3 fs accuracy. This technique bridges the gap between streaking of photoelectrons with optical lasers and deflection of relativistic electron beams with radio-frequency deflectors, and should have wide applications in many ultrashort electron beam based facilities.

physics.acc-ph↗

arXiv2019

Smart-Edge-CoCaCo: AI-Enabled Smart Edge with Joint Computation, Caching, and Communication in Heterogeneous IoT

The development of mobile communication technology, hardware, distributed computing, and artificial intelligence (AI) technology has promoted the application of edge computing in the field of heterogeneous Internet of Things (IoT). In order to overcome the defects of the traditional cloud computing model in the era of big data. In this article, we first propose a new AIenabled smart edge with heterogeneous IoT architecture which combines edge computing, caching, and communication. Then, we propose the Smart-Edge-CoCaCo algorithm. To minimize total delay and confirm the computation offloading decision, Smart-Edge-CoCaCo uses joint optimization of the wireless communication model, the collaborative filter caching model in edge cloud, and the computation offloading model. Finally, we built an emotion interaction testbed to perform computational delay experiments in real environments. The experiment results showed that the computation delay of the Smart-Edge-CoCaCo algorithm is lower than that of traditional cloud computing model with the increase of computing task data and the number of concurrent users.

cs.NI↗

arXiv2018

Urban Healthcare Big Data System Based on Crowdsourced and Cloud-Based Air Quality Indicators

The ever-accelerating process of globalization enables more than half the population to live in cities. Thus, the air quality in cities exerts critical influence on the health status of more and more urban residents. In this article, based on urban air quality data collected through meteorological sites, mobile crowdsourcing, and IoT sensing, along with users' body signals, we propose an urban healthcare big data system named UH-BigDataSys. In this article, we first introduce a method of integrating multi-source air quality data for the data preparation of artificial-intelligence-based smart urban services. Then a testbed of UH-BigDataSys is set up with the deployment of air-quality-aware healthcare applications. Finally, we provide health guidance for urban residents in aspects of respiratory diseases, outdoor travel, sleep quality, and so on. The ultimate goal of UH-BigDataSys is for urban residents to lead healthier lives.

cs.CY↗

arXiv2018

Wearable Affective Robot

With the development of the artificial intelligence (AI), the AI applications have influenced and changed people's daily life greatly. Here, a wearable affective robot that integrates the affective robot, social robot, brain wearable, and wearable 2.0 is proposed for the first time. The proposed wearable affective robot is intended for a wide population, and we believe that it can improve the human health on the spirit level, meeting the fashion requirements at the same time. In this paper, the architecture and design of an innovative wearable affective robot, which is dubbed as Fitbot, are introduced in terms of hardware and algorithm's perspectives. In addition, the important functional component of the robot-brain wearable device is introduced from the aspect of the hardware design, EEG data acquisition and analysis, user behavior perception, and algorithm deployment, etc. Then, the EEG based cognition of user's behavior is realized. Through the continuous acquisition of the in-depth, in-breadth data, the Fitbot we present can gradually enrich user's life modeling and enable the wearable robot to recognize user's intention and further understand the behavioral motivation behind the user's emotion. The learning algorithm for the life modeling embedded in Fitbot can achieve better user's experience of affective social interaction. Finally, the application service scenarios and some challenging issues of a wearable affective robot are discussed.

cs.HC↗