arXiv subjects

Dong Chen

Publications and source records attributed to Dong Chen.

At least 271 records · Page 15Linked to original sources

arXiv2020

Cross-domain Correspondence Learning for Exemplar-based Image Translation

We present a general framework for exemplar-based image translation, which synthesizes a photo-realistic image from the input in a distinct domain (e.g., semantic segmentation mask, or edge map, or pose keypoints), given an exemplar image. The output has the style (e.g., color, texture) in consistency with the semantically corresponding objects in the exemplar. We propose to jointly learn the crossdomain correspondence and the image translation, where both tasks facilitate each other and thus can be learned with weak supervision. The images from distinct domains are first aligned to an intermediate domain where dense correspondence is established. Then, the network synthesizes images based on the appearance of semantically corresponding patches in the exemplar. We demonstrate the effectiveness of our approach in several image translation tasks. Our method is superior to state-of-the-art methods in terms of image quality significantly, with the image style faithful to the exemplar with semantic consistency. Moreover, we show the utility of our method for several applications

cs.CV↗

arXiv2020

Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set

Recently, deep learning based 3D face reconstruction methods have shown promising results in both quality and efficiency.However, training deep neural networks typically requires a large volume of data, whereas face images with ground-truth 3D face shapes are scarce. In this paper, we propose a novel deep 3D face reconstruction approach that 1) leverages a robust, hybrid loss function for weakly-supervised learning which takes into account both low-level and perception-level information for supervision, and 2) performs multi-image face reconstruction by exploiting complementary information from different images for shape aggregation. Our method is fast, accurate, and robust to occlusion and large pose. We provide comprehensive experiments on three datasets, systematically comparing our method with fifteen recent methods and demonstrating its state-of-the-art performance.

cs.CV↗

arXiv2020

Curved Buildings Reconstruction from Airborne LiDAR Data by Matching and Deforming Geometric Primitives

Airborne LiDAR (Light Detection and Ranging) data is widely applied in building reconstruction, with studies reporting success in typical buildings. However, the reconstruction of curved buildings remains an open research problem. To this end, we propose a new framework for curved building reconstruction via assembling and deforming geometric primitives. The input LiDAR point cloud are first converted into contours where individual buildings are identified. After recognizing geometric units (primitives) from building contours, we get initial models by matching basic geometric primitives to these primitives. To polish assembly models, we employ a warping field for model refinements. Specifically, an embedded deformation (ED) graph is constructed via downsampling the initial model. Then, the point-to-model displacements are minimized by adjusting node parameters in the ED graph based on our objective function. The presented framework is validated on several highly curved buildings collected by various LiDAR in different cities. The experimental results, as well as accuracy comparison, demonstrate the advantage and effectiveness of our method. {The new insight attributes to an efficient reconstruction manner.} Moreover, we prove that the primitive-based framework significantly reduces the data storage to 10-20 percent of classical mesh models.

cs.CV↗

arXiv2020

Online Semantic Exploration of Indoor Maps

In this paper we propose a method to extract an abstracted floor plan from typical grid maps using Bayesian reasoning. The result of this procedure is a probabilistic generative model of the environment defined over abstract concepts. It is well suited for higher-level reasoning and communication purposes. We demonstrate the effectiveness of the approach through real-world experiments.

cs.CV↗

arXiv2020

Table-Top Scene Analysis Using Knowledge-Supervised MCMC

In this paper, we propose a probabilistic method to generate abstract scene graphs for table-top scenes from 6D object pose estimates. We explicitly make use of task-specfic context knowledge by encoding this knowledge as descriptive rules in Markov logic networks. Our approach to generate scene graphs is probabilistic: Uncertainty in the object poses is addressed by a probabilistic sensor model that is embedded in a data driven MCMC process. We apply Markov logic inference to reason about hidden objects and to detect false estimates of object poses. The effectiveness of our approach is demonstrated and evaluated in real world experiments.

cs.CV↗

arXiv2020

TCM-ICP: Transformation Compatibility Measure for Registering Multiple LIDAR Scans

Rigid registration of multi-view and multi-platform LiDAR scans is a fundamental problem in 3D mapping, robotic navigation, and large-scale urban modeling applications. Data acquisition with LiDAR sensors involves scanning multiple areas from different points of view, thus generating partially overlapping point clouds of the real world scenes. Traditionally, ICP (Iterative Closest Point) algorithm is used to register the acquired point clouds together to form a unique point cloud that captures the scanned real world scene. Conventional ICP faces local minima issues and often needs a coarse initial alignment to converge to the optimum. In this work, we present an algorithm for registering multiple, overlapping LiDAR scans. We introduce a geometric metric called Transformation Compatibility Measure (TCM) which aids in choosing the most similar point clouds for registration in each iteration of the algorithm. The LiDAR scan most similar to the reference LiDAR scan is then transformed using simplex technique. An optimization of the transformation using gradient descent and simulated annealing techniques are then applied to improve the resulting registration. We evaluate the proposed algorithm on four different real world scenes and experimental results shows that the registration performance of the proposed method is comparable or superior to the traditionally used registration methods. Further, the algorithm achieves superior registration results even when dealing with outliers.

cs.CV↗

arXiv2019

Properties of equilibrium states for geodesic flows over manifolds without focal points

We prove that for closed rank 1 manifolds without focal points the equilibrium states are unique for Hölder potentials satisfying the pressure gap condition. In addition, we provide a criterion for a continuous potential to satisfy the pressure gap condition. Moreover, we derive several ergodic properties of the unique equilibrium states including the equidistribution and the K-property.

math.DS↗

arXiv2019

Sparking mashups to form multifunctional alloy nanoparticles

Synthesizing unconventional alloys remains challenging owing to seamless interactions between kinetics and thermodynamics. High entropy alloys (HEAs), for example, draw a fundamentally new concept to enable exploring unknown regions in phase diagrams. The exploration, however, is hindered by traditional metallurgies based on liquid-solid transformation. Vapor-solid transformation that is permissible on pressure-temperature phase diagrams, offers the most kinetically efficient pathway to form any desired alloy (e.g., HEA). Here, we report that a technique called "sparking mashups", which involves a rapidly quenched vapor source and induces unrestricted mixing for alloying 55 distinct types of ultrasmall nanoparticles (NPs) with controllable compositions. Unlike the precursor feed in wet chemistry, a microseconds-long oscillatory spark controls the vapour composition, which is eventually retained in the alloy NPs. The resulting NPs range from binary to HEAs with marked thermal stability at room temperature. We show that a nanosize-effect ensures such thermal stability and mimics the role of mixing entropy in HEAs. This discovery contradicts the traditional "smaller is less stable" view while enabling the elemental combinations that have never been alloyed to date. We even break the miscibility limits by mixing bulk-immiscible systems in alloy NPs. As powerful examples, we demonstrate the alloy NPs as both high-performance fuel-cell catalysts and building blocks for three-dimensional (3D) nanoprinting to construct HEA nanostructure arrays of various architectures and compositions. Our results form the basis of new rules for guiding HEA-NP synthesis and advancing catalysis and 3D printing to new frontiers.

physics.app-ph↗

arXiv2019

Autonomous Driving using Safe Reinforcement Learning by Incorporating a Regret-based Human Lane-Changing Decision Model

It is expected that many human drivers will still prefer to drive themselves even if the self-driving technologies are ready. Therefore, human-driven vehicles and autonomous vehicles (AVs) will coexist in a mixed traffic for a long time. To enable AVs to safely and efficiently maneuver in this mixed traffic, it is critical that the AVs can understand how humans cope with risks and make driving-related decisions. On the other hand, the driving environment is highly dynamic and ever-changing, and it is thus difficult to enumerate all the scenarios and hard-code the controllers. To face up these challenges, in this work, we incorporate a human decision-making model in reinforcement learning to control AVs for safe and efficient operations. Specifically, we adapt regret theory to describe a human driver's lane-changing behavior, and fit the personalized models to individual drivers for predicting their lane-changing decisions. The predicted decisions are incorporated in the safety constraints for reinforcement learning in training and in implementation. We then use an extended version of double deep Q-network (DDQN) to train our AV controller within the safety set. By doing so, the amount of collisions in training is reduced to zero, while the training accuracy is not impinged.

cs.RO↗

arXiv2019

Superconductivity Induced at a Point Contact on the Topological Semimetal Tungsten Carbide

We report the observation of local superconductivity induced at the point contact formed between a normal metal tip and WC -- a triple point topological semimetal with super hardness. Remarkably, the maximum critical temperature is up to near 12 K but insensitive to the tip's magnetism. The lateral dimensions of the superconducting puddles were evaluated and the temperature dependencies of superconducting gap and upper critical field were also obtained. These results put constraints on the explanation of the induced superconductivity and pave a pathway for exploring topological superconductivity.

cond-mat.supr-con↗

arXiv2019

Interference mitigation techniques for a dense heterogeneous area network in machine-to-machine communications

With the advent of Machine-to-Machine (M2M) communications, various networking consumer industrial and autonomous systems exchange messages in the real world in order to achieve their objectives. Parts of these systems are comprised of short-range wireless networks in the form of clusters that collectively cover a large geographical area. In these clusters, the nodes that represent the cluster heads need to deal with two types of communications: one is within the cluster and the other is from the cluster to the sink node. As the number of clusters increases, it takes multiple hops for the cluster head to forward data to the sink node, thus resulting in a low packet delivery rate and throughput. To solve this problem, we propose a heterogeneous area network in which the cluster head is equipped with two types of radios: the IEEE 802.15.4 and IEEE 802.11 radios. The former is for the devices within the cluster to communicate, whereas the latter is for the cluster heads to communicate to the sink node. Although the IEEE 802.11 links increase the link capacity, the IEEE 802.11 radio and the IEEE 802.15.4 radio might share the 2.4 GHz unlicensed band, thus giving rise to the inter-network collisions or interference. To tackle this problem and to maintain decent Qualityof-Service (QoS) for the network, we subsequently present two interference mitigation techniques, in which a Blank Burst (BB) period is proposed so that the IEEE 802.15.4 radios can be suspended while the IEEE 802.11 radios are active. Simulation results show the proposed two methods can effectively mitigate the inter-network collisions and are superior to the existing technique, which uses an adaptive aggregation technique to mitigate the inter-network collisions.

cs.NI↗

arXiv2019

Superconducting properties in a candidate topological nodal line semimetal SnTaS$_2$ with a centrosymmetric crystal structure

We report the magnetization, electrical resistivity, specific heat measurements and band structure calculations of layered superconductor SnTaS$_2$. The experiments are performed on single crystals grown by chemical vapor transport method. The resistivity and magnetic susceptibility indicate that SnTaS$_2$ is a type-II superconductor with transition temperature $T_c = 3$ K. The upper critical field ($H_{c2}$) shows large anisotropy for magnetic field parallel to $ab$ plane ($H//ab$) and $c$ axis ($H//c$), and the temperature dependence of $H_{c2}$ for $H//ab$ shows obvious unconventional upward feature at low temperature. Band structure of SnTaS$_2$ shows several band crossings near the Fermi level, which form three nodal lines in the k$_z$ = 0 plane resulting in drumhead-like surface states when spin-orbit coupling is not considered. These results indicate that SnTaS$_2$ is a superconductor with possible topological nodal line semimetal character.

cond-mat.supr-con↗

arXiv2019

Deep Exemplar-based Video Colorization

This paper presents the first end-to-end network for exemplar-based video colorization. The main challenge is to achieve temporal consistency while remaining faithful to the reference style. To address this issue, we introduce a recurrent framework that unifies the semantic correspondence and color propagation steps. Both steps allow a provided reference image to guide the colorization of every frame, thus reducing accumulated propagation errors. Video frames are colorized in sequence based on the colorization history, and its coherency is further enforced by the temporal consistency loss. All of these components, learned end-to-end, help produce realistic videos with good temporal stability. Experiments show our result is superior to the state-of-the-art methods both quantitatively and qualitatively.

cs.CV↗

arXiv2019

Face Parsing with RoI Tanh-Warping

Face parsing computes pixel-wise label maps for different semantic components (e.g., hair, mouth, eyes) from face images. Existing face parsing literature have illustrated significant advantages by focusing on individual regions of interest (RoIs) for faces and facial components. However, the traditional crop-and-resize focusing mechanism ignores all contextual area outside the RoIs, and thus is not suitable when the component area is unpredictable, e.g. hair. Inspired by the physiological vision system of human, we propose a novel RoI Tanh-warping operator that combines the central vision and the peripheral vision together. It addresses the dilemma between a limited sized RoI for focusing and an unpredictable area of surrounding context for peripheral information. To this end, we propose a novel hybrid convolutional neural network for face parsing. It uses hierarchical local based method for inner facial components and global methods for outer facial components. The whole framework is simple and principled, and can be trained end-to-end. To facilitate future research of face parsing, we also manually relabel the training data of the HELEN dataset and will make it public. Experiments on both HELEN and LFW-PL benchmarks demonstrate that our method surpasses state-of-the-art methods.

cs.CV↗

arXiv2019

Mask-Guided Portrait Editing with Conditional GANs

Portrait editing is a popular subject in photo manipulation. The Generative Adversarial Network (GAN) advances the generating of realistic faces and allows more face editing. In this paper, we argue about three issues in existing techniques: diversity, quality, and controllability for portrait synthesis and editing. To address these issues, we propose a novel end-to-end learning framework that leverages conditional GANs guided by provided face masks for generating faces. The framework learns feature embeddings for every face component (e.g., mouth, hair, eye), separately, contributing to better correspondences for image translation, and local face editing. With the mask, our network is available to many applications, like face synthesis driven by mask, face Swap+ (including hair in swapping), and local manipulation. It can also boost the performance of face parsing a bit as an option of data augmentation.

cs.CV↗

arXiv2019

Lijiang 2.4-meter Telescope and its Instruments

Lijiang 2.4-meter Telescope(LJT), the largest common-purpose optical telescope in China, has been applied to the world-wide astronomers since 2008. It is located at Gaomeigu site, Lijiang Observatory(LJO), the southwest of China. The site has very good observational conditions. Since 10-year operation, several instruments have been equipped on the LJT. Astronomers can perform both photometric and spectral observations. The main scientific goals of LJT include photometric and spectral evolution of supernova, reverberation mapping of active galactic nucleus, physical properties of binary star and near-earth object(comet and asteroid), identification of exoplanet, and all kinds of transients. Until now, the masses of 41 high accretion rate black holes have been measured, and more than 168 supernova have been identified by the LJT. More than 190 papers related to the LJT have been published. In this paper, the general observation condition of the Gaomeigu site is introduced at first. Then, the LJT structure is described in detail, including the optical, mechanical, motion and control system. The specification of all the instruments, and some detailed parameters of the YFOSC is also presented. Finally, some important scientific results and future expectations are summarized.

astro-ph.IM↗

arXiv2019

On the $C^{\infty}$ closing lemma for Hamiltonian flows on symplectic $4$-manifolds

The main result in this paper is the $C^{\infty}$ closing lemma for a large family of Hamiltonian flows on $4$-dimensional symplectic manifolds, which includes classical Hamiltonian systems. First we prove the $C^{\infty}$ closing lemma and the $C^r$ general density theorem for geodesic flows on closed Finsler surfaces by combining a result of Asaoka-Irie with the dual lens map technique. Then we extend our results to Hamitonian flows with certain restriction. We also list some applications of our results in differential geometry and contact topology.

math.DS↗

arXiv2019

WIDER Face and Pedestrian Challenge 2018: Methods and Results

This paper presents a review of the 2018 WIDER Challenge on Face and Pedestrian. The challenge focuses on the problem of precise localization of human faces and bodies, and accurate association of identities. It comprises of three tracks: (i) WIDER Face which aims at soliciting new approaches to advance the state-of-the-art in face detection, (ii) WIDER Pedestrian which aims to find effective and efficient approaches to address the problem of pedestrian detection in unconstrained environments, and (iii) WIDER Person Search which presents an exciting challenge of searching persons across 192 movies. In total, 73 teams made valid submissions to the challenge tracks. We summarize the winning solutions for all three tracks. and present discussions on open problems and potential research directions in these topics.

cs.CV↗