Search arXiv⌕ Search

arXiv subjects

Ming Liu

Publications and source records attributed to Ming Liu.

At least 487 records · Page 27Linked to original sources
arXiv2018

Learning Warped Guidance for Blind Face Restoration

This paper studies the problem of blind face restoration from an unconstrained blurry, noisy, low-resolution, or compressed image (i.e., degraded observation). For better recovery of fine facial details, we modify the problem setting by taking both the degraded observation and a high-quality guided image of the same identity as input to our guided face restoration network (GFRNet). However, the degraded observation and guided image generally are different in pose, illumination and expression, thereby making plain CNNs (e.g., U-Net) fail to recover fine and identity-aware facial details. To tackle this issue, our GFRNet model includes both a warping subnetwork (WarpNet) and a reconstruction subnetwork (RecNet). The WarpNet is introduced to predict flow field for warping the guided image to correct pose and expression (i.e., warped guidance), while the RecNet takes the degraded observation and warped guidance as input to produce the restoration result. Due to that the ground-truth flow field is unavailable, landmark loss together with total variation regularization are incorporated to guide the learning of WarpNet. Furthermore, to make the model applicable to blind restoration, our GFRNet is trained on the synthetic data with versatile settings on blur kernel, noise level, downsampling scale factor, and JPEG quality factor. Experiments show that our GFRNet not only performs favorably against the state-of-the-art image and face restoration methods, but also generates visually photo-realistic results on real degraded facial images.

cs.CV↗
arXiv2018

RRT* Combined with GVO for Real-time Nonholonomic Robot Navigation in Dynamic Environment

Challenges persist in nonholonomic robot navigation in dynamic environments. This paper presents a framework for such navigation based on the model of generalized velocity obstacles (GVO). The idea of velocity obstacles has been well studied and developed for obstacle avoidance since being proposed in 1998. Though it has been proved to be successful, most studies have assumed equations of motion to be linear, which limits their application to holonomic robots. In addition, more attention has been paid to the immediate reaction of robots, while advance planning has been neglected. By applying the GVO model to differential drive robots and by combining it with RRT*, we reduce the uncertainty of the robot trajectory, thus further reducing the range of concern, and save both computation time and running time. By introducing uncertainty for the dynamic obstacles with a Kalman filter, we dilute the risk of considering the obstacles as uniformly moving along a straight line and guarantee the safety. Special concern is given to path generation, including curvature check, making the generated path feasible for nonholonomic robots. We experimentally demonstrate the feasibility of the framework.

cs.RO↗
arXiv2018

A Survey of Deep Network Solutions for Learning Control in Robotics: From Reinforcement to Imitation

Deep learning techniques have been widely applied, achieving state-of-the-art results in various fields of study. This survey focuses on deep learning solutions that target learning control policies for robotics applications. We carry out our discussions on the two main paradigms for learning control with deep networks: deep reinforcement learning and imitation learning. For deep reinforcement learning (DRL), we begin from traditional reinforcement learning algorithms, showing how they are extended to the deep context and effective mechanisms that could be added on top of the DRL algorithms. We then introduce representative works that utilize DRL to solve navigation and manipulation tasks in robotics. We continue our discussion on methods addressing the challenge of the reality gap for transferring DRL policies trained in simulation to real-world scenarios, and summarize robotics simulation platforms for conducting DRL research. For imitation leaning, we go through its three main categories, behavior cloning, inverse reinforcement learning and generative adversarial imitation learning, by introducing their formulations and their corresponding robotics applications. Finally, we discuss the open challenges and research frontiers.

cs.RO↗
arXiv2018

Solution-processed ZnO as the efficient passivation and electron selective layer of silicon solar cells

Solution-processed intrinsic ZnO and Al doped ZnO (ZnO:Al) were spin coated on textured n-type c-Si wafer to replace the phosphorus doped amorphous silicon as the electron selective transport layer (ESTL) of the Si heterojunction (SHJ) solar cells. Besides the function of electron selective transportation, the non-doped ZnO was found to possess certain passivation effect on c-Si wafer. The SHJ solar cells with different combinations of passivation layer (intrinsic a-Si:H, SiOx and non-doped ZnO) and electron transport layer (non-doped ZnO and ZnO:Al ) were fabricated and compared. An efficiency up to 18.46% was achieved on a SHJ solar cell with an a-Si:H/ZnO:Al double layer back structure. And, the all solution-processed non-doped ZnO/ZnO:Al combination layer presents fairly good electron selective transportation property for SHJ solar cell, resulting in an efficiency of 17.13%. The carrier transport based on energy band diagrams of the rear side of the solar cells has been discussed related to the performance of the SHJ solar cells.

physics.app-ph↗
arXiv2018

Ionic Modulation of Interfacial Magnetism through Electrostatic Doping in Pt/YIG bilayer heterostructure

Voltage modulation of yttrium iron garnet (YIG) with compactness, high speed response, energy efficiency and both practical/theoretical siginificances can be widely applied to various YIG based spintronics such as spin Hall, spin pumping, spin Seeback effects. Here we initial an ionic modulation of interfacial magnetism process on YIG/Pt bilayer heterostructures, where the Pt capping would influence the ferromagnetic (FMR) field position significantly, and realize a significant magnetism enhancement in bilayer system. A large voltage induced FMR field shifts of 690 Oe has been achieved in YIG (13 nm)/Pt (3 nm) multilayer heterostructures under a small voltage bias of 4.5 V. The remarkable ME tunability comes from voltage induced extra FM ordering in Pt metal layer near the Pt/YIG interface. The first-principle theoretical simulation reveal that the electrostatic doping induced Pt5+ ions have strong magnetic ordering due to uncompensated d orbit electrons. The large voltage control of FMR change pave a foundation towards novel voltage tunable YIG based spintronics.

cond-mat.mtrl-sci↗
arXiv2018

Socially Compliant Navigation through Raw Depth Inputs with Generative Adversarial Imitation Learning

We present an approach for mobile robots to learn to navigate in dynamic environments with pedestrians via raw depth inputs, in a socially compliant manner. To achieve this, we adopt a generative adversarial imitation learning (GAIL) strategy, which improves upon a pre-trained behavior cloning policy. Our approach overcomes the disadvantages of previous methods, as they heavily depend on the full knowledge of the location and velocity information of nearby pedestrians, which not only requires specific sensors, but also the extraction of such state information from raw sensory input could consume much computation time. In this paper, our proposed GAIL-based model performs directly on raw depth inputs and plans in real-time. Experiments show that our GAIL-based approach greatly improves the safety and efficiency of the behavior of mobile robots from pure behavior cloning. The real-world deployment also shows that our method is capable of guiding autonomous vehicles to navigate in a socially compliant manner directly through raw depth inputs. In addition, we release a simulation plugin for modeling pedestrian behaviors based on the social force model.

cs.RO↗
arXiv2018

On fusion procedure for the two-parameter quantum algebra in type A

Finite dimensional irreducible modules of the two-parameter quantum enveloping algebra $U_{r,s}(\mathfrak{sl}_n)$ are explicitly constructed using the fusion procedure when $rs^{-1}$ is generic. This provides an alternative and combinatorial description of the Schur-Weyl duality for the two-parameter quantum linear algebras of type $A$.

math.QA↗
arXiv2017

Calculating Semantic Similarity between Academic Articles using Topic Event and Ontology

Determining semantic similarity between academic documents is crucial to many tasks such as plagiarism detection, automatic technical survey and semantic search. Current studies mostly focus on semantic similarity between concepts, sentences and short text fragments. However, document-level semantic matching is still based on statistical information in surface level, neglecting article structures and global semantic meanings, which may cause the deviation in document understanding. In this paper, we focus on the document-level semantic similarity issue for academic literatures with a novel method. We represent academic articles with topic events that utilize multiple information profiles, such as research purposes, methodologies and domains to integrally describe the research work, and calculate the similarity between topic events based on the domain ontology to acquire the semantic similarity between articles. Experiments show that our approach achieves significant performance compared to state-of-the-art methods.

cs.CL↗
arXiv2017

LiDAR and Inertial Fusion for Pose Estimation by Non-linear Optimization

Pose estimation purely based on 3D point-cloud could suffer from degradation, e.g. scan blocks or scans in repetitive environments. To deal with this problem, we propose an approach for fusing 3D spinning LiDAR and IMU to estimate the ego-motion of the sensor body. The main idea of our work is to optimize the poses and states of two kinds of sensors with non-linear optimization methods. On the one hand, a bunch of IMU measurements are considered as a relative constraint using pre-integration and the state errors can be minimized with the help of laser pose estimation and non-linear optimization algorithms; on the other hand, the optimized IMU pose outputs can provide a better initial for the subsequent point-cloud matching. The method is evaluated under both simulation and real tests with comparison to the state-of-the-art. The results show that the proposed method can provide better pose estimation performance even in the degradation cases.

cs.RO↗
arXiv2017

Towards continuous control of flippers for a multi-terrain robot using deep reinforcement learning

In this paper we focus on developing a control algorithm for multi-terrain tracked robots with flippers using a reinforcement learning (RL) approach. The work is based on the deep deterministic policy gradient (DDPG) algorithm, proven to be very successful in simple simulation environments. The algorithm works in an end-to-end fashion in order to control the continuous position of the flippers. This end-to-end approach makes it easy to apply the controller to a wide array of circumstances, but the huge flexibility comes to the cost of an increased difficulty of solution. The complexity of the task is enlarged even more by the fact that real multi-terrain robots move in partially observable environments. Notwithstanding these complications, being able to smoothly control a multi-terrain robot can produce huge benefits in impaired people daily lives or in search and rescue situations.

cs.RO↗
arXiv2017

Characterization of a RS-LiDAR for 3D Perception

High precision 3D LiDARs are still expensive and hard to acquire. This paper presents the characteristics of RS-LiDAR, a model of low-cost LiDAR with sufficient supplies, in comparison with VLP-16. The paper also provides a set of evaluations to analyze the characterizations and performances of LiDARs sensors. This work analyzes multiple properties, such as drift effects, distance effects, color effects and sensor orientation effects, in the context of 3D perception. By comparing with Velodyne LiDAR, we found RS-LiDAR as a cheaper and acquirable substitute of VLP-16 with similar efficiency.

cs.RO↗
arXiv2017

Plugo: a VLC Systematic Perspective of Large-scale Indoor Localization

Indoor localization based on Visible Light Communication (VLC) has been in favor with both the academia and industry for years. In this paper, we present a prototyping photodiode-based VLC system towards large-scale localization. Specially, we give in-depth analysis of the design constraints and considerations for large-scale indoor localization research. After that we identify the key enablers for such systems: 1) distributed architecture, 2) one-way communication, and 3) random multiple access. Accordingly, we propose Plugo -- a photodiode-based VLC system conforming to the aforementioned criteria. We present a compact design of the VLC-compatible LED bulbs featuring plug-and-go use-cases. The basic framed slotted Additive Links On-line Hawaii Area (ALOHA) is exploited to achieve random multiple access over the shared optical medium. We show its effectiveness in beacon broadcasting by experiments, and further demonstrate its scalability to large-scale scenarios through simulations. Finally, preliminary localization experiments are conducted using fingerprinting-based methods in a customized testbed, achieving an average accuracy of 0.14 m along with a 90-percentile accuracy of 0.33 m.

cs.NI↗
arXiv2017

Virtual-to-real Deep Reinforcement Learning: Continuous Control of Mobile Robots for Mapless Navigation

We present a learning-based mapless motion planner by taking the sparse 10-dimensional range findings and the target position with respect to the mobile robot coordinate frame as input and the continuous steering commands as output. Traditional motion planners for mobile ground robots with a laser range sensor mostly depend on the obstacle map of the navigation environment where both the highly precise laser sensor and the obstacle map building work of the environment are indispensable. We show that, through an asynchronous deep reinforcement learning method, a mapless motion planner can be trained end-to-end without any manually designed features and prior demonstrations. The trained planner can be directly applied in unseen virtual and real environments. The experiments show that the proposed mapless motion planner can navigate the nonholonomic mobile robot to the desired targets without colliding with any obstacles.

cs.RO↗
arXiv2017

PCA-aided Fully Convolutional Networks for Semantic Segmentation of Multi-channel fMRI

Semantic segmentation of functional magnetic resonance imaging (fMRI) makes great sense for pathology diagnosis and decision system of medical robots. The multi-channel fMRI provides more information of the pathological features. But the increased amount of data causes complexity in feature detections. This paper proposes a principal component analysis (PCA)-aided fully convolutional network to particularly deal with multi-channel fMRI. We transfer the learned weights of contemporary classification networks to the segmentation task by fine-tuning. The results of the convolutional network are compared with various methods e.g. k-NN. A new labeling strategy is proposed to solve the semantic segmentation problem with unclear boundaries. Even with a small-sized training dataset, the test results demonstrate that our model outperforms other pathological feature detection methods. Besides, its forward inference only takes 90 milliseconds for a single set of fMRI data. To our knowledge, this is the first time to realize pixel-wise labeling of multi-channel magnetic resonance image using FCN.

cs.CV↗
arXiv2017

Experiments on bright field and dark field high energy electron imaging with thick target material

Using a high energy electron beam for the imaging of high density matter with both high spatial-temporal and areal density resolution under extreme states of temperature and pressure is one of the critical challenges in high energy density physics . When a charged particle beam passes through an opaque target, the beam will be scattered with a distribution that depends on the thickness of the material. By collecting the scattered beam either near or off axis, so-called bright field or dark field images can be obtained. Here we report on an electron radiography experiment using 45 MeV electrons from an S-band photo-injector, where scattered electrons, after interacting with a sample, are collected and imaged by a quadrupole imaging system. We achieved a few micrometers (about 4 micrometers) spatial resolution and about 10 micrometers thickness resolution for a silicon target of 300-600 micron thickness. With addition of dark field images that are captured by selecting electrons with large scattering angle, we show that more useful information in determining external details such as outlines, boundaries and defects can be obtained.

physics.acc-ph↗
arXiv2017

Extrinsic Calibration of 3D Range Finder and Camera without Auxiliary Object or Human Intervention

Fusion of heterogeneous extroceptive sensors is the most effient and effective way to representing the environment precisely, as it overcomes various defects of each homogeneous sensor. The rigid transformation (aka. extrinsic parameters) of heterogeneous sensory systems should be available before precisely fusing the multisensor information. Researchers have proposed several approaches to estimating the extrinsic parameters. These approaches require either auxiliary objects, like chessboards, or extra help from human to select correspondences. In this paper, we proposed a novel extrinsic calibration approach for the extrinsic calibration of range and image sensors. As far as we know, it is the first automatic approach with no requirement of auxiliary objects or any human interventions. First, we estimate the initial extrinsic parameters from the individual motion of the range finder and the camera. Then we extract lines in the image and point-cloud pairs, to refine the line feature associations by the initial extrinsic parameters. At the end, we discussed the degenerate case which may lead to the algorithm failure and validate our approach by simulation. The results indicate high-precision extrinsic calibration results against the ground-truth.

cs.CV↗
arXiv2017

A review for compact model of graphene field-effect transistors

Graphene has attracted enormous interests due to its unique physical, mechanical, and electrical properties. Specially, graphene-based field-effect transistors (FETs) have evolved rapidly and are now considered as an option for conventional silicon devices. As a critical step in the design cycle of modern IC products, compact model refers to the development of models for integrated semiconductor devices for use in circuit simulations. The purpose of this review is to provide a theoretical description of current compact model of graphene field-effect transistors. Special attention is devoted to the charge sheet model, drift-diffusion model, Boltzmann equation, density of states (DOS), and surface-potential-based compact model. Finally, an outlook of this field is briefly discussed.

cond-mat.mes-hall↗