arXiv subjects

Qian Zhang

Publications and source records attributed to Qian Zhang.

At least 397 records · Page 22Linked to original sources

arXiv2020

Fast Neural Network Adaptation via Parameter Remapping and Architecture Search

Deep neural networks achieve remarkable performance in many computer vision tasks. Most state-of-the-art (SOTA) semantic segmentation and object detection approaches reuse neural network architectures designed for image classification as the backbone, commonly pre-trained on ImageNet. However, performance gains can be achieved by designing network architectures specifically for detection and segmentation, as shown by recent neural architecture search (NAS) research for detection and segmentation. One major challenge though, is that ImageNet pre-training of the search space representation (a.k.a. super network) or the searched networks incurs huge computational cost. In this paper, we propose a Fast Neural Network Adaptation (FNA) method, which can adapt both the architecture and parameters of a seed network (e.g. a high performing manually designed backbone) to become a network with different depth, width, or kernels via a Parameter Remapping technique, making it possible to utilize NAS for detection/segmentation tasks a lot more efficiently. In our experiments, we conduct FNA on MobileNetV2 to obtain new networks for both segmentation and detection that clearly out-perform existing networks designed both manually and by NAS. The total computation cost of FNA is significantly less than SOTA segmentation/detection NAS approaches: 1737$\times$ less than DPC, 6.8$\times$ less than Auto-DeepLab and 7.4$\times$ less than DetNAS. The code is available at https://github.com/JaminFong/FNA.

cs.CV↗

arXiv2020

Cardiovascular risk and work stress in biomedical researchers in China: An observational, big data study protocol

Introduction: Internet technologies could strengthen data collection and integration and have been used extensively in public health research. It is necessary to apply this technology to further investigate the behaviour and health of biomedical researchers. A browser-based extension was developed by researchers and clinicians to promote the collection and analysis of researchers' behavioural and psychological data. This protocol illustrates an observational study aimed at (1) characterising the health status of biomedical researchers in China and assessing work stress, job satisfaction, role conflict, role ambiguity, and family support; (2) identifying the association between work, behaviour, and health; and (3) investigating the association between behaviour and mental status. Our findings will contribute to the understanding of the influences of job, work environment, and family support on the mental and physical health of biomedical researchers. Methods and analysis: This is a prospective observational study; all candidates will be recruited from China. Participants will install an extension on their Internet browsers, which will collect data when they are accessing PubMed. A web-based survey will be sent to the user interfaces every 6 months that will involve sociodemographic variables, perceived stress scale, job satisfaction scale, role conflict and ambiguity scale, and family support scale. Machine-learning algorithms will analyse the data generated during daily access. Ethics and dissemination: This study received ethical approval from the ethics committee of the Shanghai Children's Medical Centre (reference number SCMCIRB-K2018082). Study results will be disseminated through peer-reviewed publications and conference presentations.

cs.CY↗

arXiv2020

Boundary Guidance Hierarchical Network for Real-Time Tongue Segmentation

Automated tongue image segmentation in tongue images is a challenging task for two reasons: 1) there are many pathological details on the tongue surface, which affect the extraction of the boundary; 2) the shapes of the tongues captured from various persons (with different diseases) are quite different. To deal with the challenge, a novel end-to-end Boundary Guidance Hierarchical Network (BGHNet) with a new hybrid loss is proposed in this paper. In the new approach, firstly Context Feature Encoder Module (CFEM) is built upon the bottomup pathway to confront with the shrinkage of the receptive field. Secondly, a novel hierarchical recurrent feature fusion module (HRFFM) is adopt to progressively and hierarchically refine object maps to recover image details by integrating local context information. Finally, the proposed hybrid loss in a four hierarchy-pixel, patch, map and boundary guides the network to effectively segment the tongue regions and accurate tongue boundaries. BGHNet is applied to a set of tongue images. The experimental results suggest that the proposed approach can achieve the latest tongue segmentation performance. And in the meantime, the lightweight network contains only 15.45M parameters and performs only 11.22GFLOPS.

eess.IV↗

arXiv2020

Active Lighting Recurrence by Parallel Lighting Analogy for Fine-Grained Change Detection

This paper studies a new problem, namely active lighting recurrence (ALR) that physically relocalizes a light source to reproduce the lighting condition from single reference image for a same scene, which may suffer from fine-grained changes during twice observations. ALR is of great importance for fine-grained visual inspection and change detection, because some phenomena or minute changes can only be clearly observed under particular lighting conditions. Therefore, effective ALR should be able to online navigate a light source toward the target pose, which is challenging due to the complexity and diversity of real-world lighting and imaging processes. To this end, we propose to use the simple parallel lighting as an analogy model and based on Lambertian law to compose an instant navigation ball for this purpose. We theoretically prove the feasibility, i.e., equivalence and convergence, of this ALR approach for realistic near point light source and small near surface light source. Besides, we also theoretically prove the invariance of our ALR approach to the ambiguity of normal and lighting decomposition. The effectiveness and superiority of the proposed approach have been verified by both extensive quantitative experiments and challenging real-world tasks on fine-grained change detection of cultural heritages. We also validate the generality of our approach to non-Lambertian scenes.

cs.CV↗

arXiv2020

Enabling Low-Power OFDM for IoT by Exploiting Asymmetric Clock Rates

The conventional high-speed Wi-Fi has recently become a contender for low-power Internet-of-Things (IoT) communications. OFDM continues its adoption in the new IoT Wi-Fi standard due to its spectrum efficiency that can support the demand of massive IoT connectivity. While the IoT Wi-Fi standard offers many new features to improve power and spectrum efficiency, the basic physical layer (PHY) structure of transceiver design still conforms to its conventional design rationale where access points (AP) and clients employ the same OFDM PHY. In this paper, we argue that current Wi-Fi PHY design does not take full advantage of the inherent asymmetry between AP and IoT. To fill the gap, we propose an asymmetric design where IoT devices transmit uplink packets using the lowest power while pushing all the decoding burdens to the AP side. Such a design utilizes the sufficient power and computational resources at AP to trade for the transmission (TX) power of IoT devices. The core technique enabling this asymmetric design is that the AP takes full power of its high clock rate to boost the decoding ability. We provide an implementation of our design and show that it can reduce up to 88% of the IoT's TX power when the AP sets $8\times$ clock rate.

cs.NI↗

arXiv2020

Transformer Transducer: A Streamable Speech Recognition Model with Transformer Encoders and RNN-T Loss

In this paper we present an end-to-end speech recognition model with Transformer encoders that can be used in a streaming speech recognition system. Transformer computation blocks based on self-attention are used to encode both audio and label sequences independently. The activations from both audio and label encoders are combined with a feed-forward layer to compute a probability distribution over the label space for every combination of acoustic frame position and label history. This is similar to the Recurrent Neural Network Transducer (RNN-T) model, which uses RNNs for information encoding instead of Transformer encoders. The model is trained with the RNN-T loss well-suited to streaming decoding. We present results on the LibriSpeech dataset showing that limiting the left context for self-attention in the Transformer layers makes decoding computationally tractable for streaming, with only a slight degradation in accuracy. We also show that the full attention version of our model beats the-state-of-the art accuracy on the LibriSpeech benchmarks. Our results also show that we can bridge the gap between full attention and limited attention versions of our model by attending to a limited number of future frames.

eess.AS↗

arXiv2020

RepChain: A Reputation-based Secure, Fast and High Incentive Blockchain System via Sharding

In today's blockchain system, designing a secure and high throughput blockchain on par with a centralized payment system is a difficult task. Sharding is one of the most worthwhile emerging technologies for improving the system throughput while maintain high security level. However, previous sharding related designs have two main limitations: Firstly, the throughput of their random-based sharding system is not high enough as they did not leverage the heterogeneity among validators. Secondly, to design an incentive mechanism to promote cooperation could incur a huge overhead on their system. In this paper, we propose RepChain, a reputation-based secure and fast blockchain system via sharding, which also provides high incentive to stimulate node cooperation. RepChain utilizes reputation to explicitly characterize the heterogeneity among the validators and lay the foundation for the incentive mechanism. We propose a new double-chain architecture which includes transaction chain and reputation chain. For transaction chain, a Raft-based synchronous consensus that can achieve high throughput has been presented. For reputation chain, the synchronous Byzantine fault tolerance that combines collective signing has been utilized to achieve a consensus on both reputation score and the related transaction blocks. It supports a high throughput transaction chain with moderate generation speed. Moreover, we propose a reputation-based sharding and leader selection scheme. To analyze the security of RepChain, we propose a recursive formula to calculate the epoch security within only O(km^2) time. Furthermore, we implement and evaluate RepChain on the Amazon Web Service platform. The results show our solution can enhance both throughout and security level of the existing sharding-based blockchain system.

cs.CR↗

arXiv2020

FasterSeg: Searching for Faster Real-time Semantic Segmentation

We present FasterSeg, an automatically designed semantic segmentation network with not only state-of-the-art performance but also faster speed than current methods. Utilizing neural architecture search (NAS), FasterSeg is discovered from a novel and broader search space integrating multi-resolution branches, that has been recently found to be vital in manually designed segmentation models. To better calibrate the balance between the goals of high accuracy and low latency, we propose a decoupled and fine-grained latency regularization, that effectively overcomes our observed phenomenons that the searched networks are prone to "collapsing" to low-latency yet poor-accuracy models. Moreover, we seamlessly extend FasterSeg to a new collaborative search (co-searching) framework, simultaneously searching for a teacher and a student network in the same single run. The teacher-student distillation further boosts the student model's accuracy. Experiments on popular segmentation benchmarks demonstrate the competency of FasterSeg. For example, FasterSeg can run over 30% faster than the closest manually designed competitor on Cityscapes, while maintaining comparable accuracy.

cs.CV↗

arXiv2020

ECGadv: Generating Adversarial Electrocardiogram to Misguide Arrhythmia Classification System

Deep neural networks (DNNs)-powered Electrocardiogram (ECG) diagnosis systems recently achieve promising progress to take over tedious examinations by cardiologists. However, their vulnerability to adversarial attacks still lack comprehensive investigation. The existing attacks in image domain could not be directly applicable due to the distinct properties of ECGs in visualization and dynamic properties. Thus, this paper takes a step to thoroughly explore adversarial attacks on the DNN-powered ECG diagnosis system. We analyze the properties of ECGs to design effective attacks schemes under two attacks models respectively. Our results demonstrate the blind spots of DNN-powered diagnosis systems under adversarial attacks, which calls attention to adequate countermeasures.

cs.LG↗

arXiv2019

Lightweight Sybil-Resilient Multi-Robot Networks by Multipath Manipulation

Wireless networking opens up many opportunities to facilitate miniaturized robots in collaborative tasks, while the openness of wireless medium exposes robots to the threats of Sybil attackers, who can break the fundamental trust assumption in robotic collaboration by forging a large number of fictitious robots. Recent advances advocate the adoption of bulky multi-antenna systems to passively obtain fine-grained physical layer signatures, rendering them unaffordable to miniaturized robots. To overcome this conundrum, this paper presents ScatterID, a lightweight system that attaches featherlight and batteryless backscatter tags to single-antenna robots to defend against Sybil attacks. Instead of passively "observing" signatures, ScatterID actively "manipulates" multipath propagation by using backscatter tags to intentionally create rich multipath features obtainable to a single-antenna robot. These features are used to construct a distinct profile to detect the real signal source, even when the attacker is mobile and power-scaling. We implement ScatterID on the iRobot Create platform and evaluate it in typical indoor and outdoor environments. The experimental results show that our system achieves a high AUROC of 0.988 and an overall accuracy of 96.4% for identity verification.

cs.CR↗

arXiv2019

AugFPN: Improving Multi-scale Feature Learning for Object Detection

Current state-of-the-art detectors typically exploit feature pyramid to detect objects at different scales. Among them, FPN is one of the representative works that build a feature pyramid by multi-scale features summation. However, the design defects behind prevent the multi-scale features from being fully exploited. In this paper, we begin by first analyzing the design defects of feature pyramid in FPN, and then introduce a new feature pyramid architecture named AugFPN to address these problems. Specifically, AugFPN consists of three components: Consistent Supervision, Residual Feature Augmentation, and Soft RoI Selection. AugFPN narrows the semantic gaps between features of different scales before feature fusion through Consistent Supervision. In feature fusion, ratio-invariant context information is extracted by Residual Feature Augmentation to reduce the information loss of feature map at the highest pyramid level. Finally, Soft RoI Selection is employed to learn a better RoI feature adaptively after feature fusion. By replacing FPN with AugFPN in Faster R-CNN, our models achieve 2.3 and 1.6 points higher Average Precision (AP) when using ResNet50 and MobileNet-v2 as backbone respectively. Furthermore, AugFPN improves RetinaNet by 1.6 points AP and FCOS by 0.9 points AP when using ResNet50 as backbone. Codes will be made available.

cs.CV↗

arXiv2019

VarGFaceNet: An Efficient Variable Group Convolutional Neural Network for Lightweight Face Recognition

To improve the discriminative and generalization ability of lightweight network for face recognition, we propose an efficient variable group convolutional network called VarGFaceNet. Variable group convolution is introduced by VarGNet to solve the conflict between small computational cost and the unbalance of computational intensity inside a block. We employ variable group convolution to design our network which can support large scale face identification while reduce computational cost and parameters. Specifically, we use a head setting to reserve essential information at the start of the network and propose a particular embedding setting to reduce parameters of fully-connected layer for embedding. To enhance interpretation ability, we employ an equivalence of angular distillation loss to guide our lightweight network and we apply recursive knowledge distillation to relieve the discrepancy between the teacher model and the student model. The champion of deepglint-light track of LFR (2019) challenge demonstrates the effectiveness of our model and approach. Implementation of VarGFaceNet will be released at https://github.com/zma-c-137/VarGFaceNet soon.

cs.CV↗

arXiv2019

Sub-cycle coherent control of ionic dynamics via transient ionization injection

We investigate the interwoven dynamic evolutions of neutral nitrogen molecules together with nitrogen ions created through transient tunnel ionization in an intense laser field. By treating the molecules as open quantum systems, it is found that considering real-time injection of ions and strong couplings among their electronic states, nitrogen molecular ions are primarily populated in the electronically excited states, rather than staying in the ground state as predicted by the well-known tunneling theory. The unexpected result is attributed to sub-cycle switch-on of time-dependent polarization by transient ionization and dynamic Stark shift mediated near-resonant multiphoton transitions. Their combined contribution also causes that the vibrational distribution of N$_2^+$ does not comply with Franck-Condon principle. These findings corroborate the mechanism of nitrogen molecular ion lasing and are likely to be universal. The present work opens a new route to explore the important role of transient ionization injection in strong-field induced non-equilibrium dynamics.

physics.atom-ph↗

arXiv2019

State-Aware Rate Adaptation for UAVs by Incorporating On-Board Sensors

Nowadays unmanned aerial vehicles (UAVs) are being widely applied to a wealth of civil and military applications. Robust and high-throughput wireless communication is the crux of these UAV applications. Yet, air-to-ground links suffer from time-varying channels induced by the agile mobility and dynamic environments. Rate adaptation algorithms are generally used to choose the optimal data rate based on the current channel conditions. State-of-the-art approaches leverage physical layer information for rate adaptation, and they work well under certain conditions. However, the above protocols still have limitation under constantly changing flight states and environments for air-to-ground links. To solve this problem, we propose StateRate, a state-optimized rate adaptation algorithm that fully exploits the characteristics of UAV systems using a hybrid deep learning model. The key observation is that the rate adaptation strategy needs to be adjusted according to motion-dependent channel models, which can be reflected by flight states. In this work, the rate adaptation protocol is enhanced with the help of the on-board sensors in UAVs. To make full use of the sensor data, we introduce a learning-based prediction module by leveraging the internal state to dynamically store temporal features under variable flight states. We also present an online learning algorithm by employing the pre-trained model that adapts the rate adaptation algorithm to different environments. We implement our algorithm on a commercial UAV platform and evaluate it in various environments. The results demonstrate that our system outperforms the best-known rate adaptation algorithm up to 53% in terms of throughput when the velocity is 2-6~m/s.

cs.NI↗

arXiv2019

Sensor-Augmented Neural Adaptive Bitrate Video Streaming on UAVs

Recent advances in unmanned aerial vehicle (UAV) technology have revolutionized a broad class of civil and military applications. However, the designs of wireless technologies that enable real-time streaming of high-definition video between UAVs and ground clients present a conundrum. Most existing adaptive bitrate (ABR) algorithms are not optimized for the air-to-ground links, which usually fluctuate dramatically due to the dynamic flight states of the UAV. In this paper, we present SA-ABR, a new sensor-augmented system that generates ABR video streaming algorithms with the assistance of various kinds of inherent sensor data that are used to pilot UAVs. By incorporating the inherent sensor data with network observations, SA-ABR trains a deep reinforcement learning (DRL) model to extract salient features from the flight state information and automatically learn an ABR algorithm to adapt to the varying UAV channel capacity through the training process. SA-ABR does not rely on any assumptions or models about UAV's flight states or the environment, but instead, it makes decisions by exploiting temporal properties of past throughput through the long short-term memory (LSTM) to adapt itself to a wide range of highly dynamic environments. We have implemented SA-ABR in a commercial UAV and evaluated it in the wild. We compare SA-ABR with a variety of existing state-of-the-art ABR algorithms, and the results show that our system outperforms the best known existing ABR algorithm by 21.4% in terms of the average quality of experience (QoE) reward.

eess.SP↗

arXiv2019

Progressive Sparse Local Attention for Video object detection

Transferring image-based object detectors to the domain of videos remains a challenging problem. Previous efforts mostly exploit optical flow to propagate features across frames, aiming to achieve a good trade-off between accuracy and efficiency. However, introducing an extra model to estimate optical flow can significantly increase the overall model size. The gap between optical flow and high-level features can also hinder it from establishing spatial correspondence accurately. Instead of relying on optical flow, this paper proposes a novel module called Progressive Sparse Local Attention (PSLA), which establishes the spatial correspondence between features across frames in a local region with progressively sparser stride and uses the correspondence to propagate features. Based on PSLA, Recursive Feature Updating (RFU) and Dense Feature Transforming (DenseFT) are proposed to model temporal appearance and enrich feature representation respectively in a novel video object detection framework. Experiments on ImageNet VID show that our method achieves the best accuracy compared to existing methods with smaller model size and acceptable runtime speed.

cs.CV↗

arXiv2019

Improved Mix-up with KL-Entropy for Learning From Noisy Labels

Despite the deep neural networks (DNN) has achieved excellent performance in image classification researches, the training of DNNs needs a large of clean data with accurate annotations. The collect of a dataset is easy, but it is difficult to annotate the collecting data. On the websites, there exist a lot of image data which contains inaccurate annotations, but training on these datasets may make networks easier to over-fit the noisy labels and cause performance degradation. In this work, we propose an improved joint optimization framework, which mixed the mix-up entropy and Kullback-Leibler (KL) entropy as the loss function. The new loss function can give the better fine-tuning after the framework updates both the label annotations. We conduct experiments on CIFAR-10 dataset and Clothing1M dataset. The result shows the advantageous performance of our approach compared with other state-of-the-art methods.

cs.CV↗

arXiv2019

Discontinuous Galerkin methods for the Ostrovsky-Vakhnenko equation

In this paper, we develop discontinuous Galerkin (DG) methods for the Ostrovsky-Vakhnenko (OV) equation, which yields the shock solutions and singular soliton solutions, such as peakon, cuspon and loop solitons. The OV equation has also been shown to have a bi-Hamiltonian structure. We directly develop the energy stable or Hamiltonian conservative discontinuous Galerkin (DG) schemes for the OV equation. Error estimates for the two energy stable schemes are also proved. For some singular solutions, including cuspon and loop soliton solutions, the hodograph transformation is adopted to transform the OV equation or the generalized OV system to the coupled dispersionless (CD) system. Subsequently, two DG schemes are constructed for the transformed CD system. Numerical experiments are provided to demonstrate the accuracy and capability of the DG schemes, including shock solution and, peakon, cuspon and loop soliton solutions.

math.NA↗