arXiv subjects

Li Zhu

Publications and source records attributed to Li Zhu.

At least 109 records · Page 6Linked to original sources

arXiv2021

Fermi level equilibration at the metal-molecule interface in plasmonic systems

We highlight a new metal-molecule charge transfer process by tuning the Fermi energy of plasmonic silver nanoparticles (AgNPs) in-situ. The strong adsorption of halide ions upshifts the Fermi level of AgNPs by up to ~0.3 eV in the order: Cl-<Br-<I-, favoring the spontaneous charge transfer to aligned molecular acceptor orbitals until charge neutrality across the interface is achieved. By carefully quantifying experimentally and theoretically the Fermi level upshift, we show for the first time that this effect is comparable in energy to different plasmonic effects such as the plasmoelectric effect or hot-carriers production. Moreover, by monitoring in-situ the adsorption dynamic of halide ions in different AgNP-molecule systems, we show for the first time that the catalytic role of halide ions in plasmonic nanostructures depends on the surface affinity of halide ions compared to that of the target molecule.

physics.chem-ph↗

arXiv2021

Detecting Small Objects in Thermal Images Using Single-Shot Detector

SSD (Single Shot Multibox Detector) is one of the most successful object detectors for its high accuracy and fast speed. However, the features from shallow layer (mainly Conv4_3) of SSD lack semantic information, resulting in poor performance in small objects. In this paper, we proposed DDSSD (Dilation and Deconvolution Single Shot Multibox Detector), an enhanced SSD with a novel feature fusion module which can improve the performance over SSD for small object detection. In the feature fusion module, dilation convolution module is utilized to enlarge the receptive field of features from shallow layer and deconvolution module is adopted to increase the size of feature maps from high layer. Our network achieves 79.7% mAP on PASCAL VOC2007 test and 28.3% mmAP on MS COCO test-dev at 41 FPS with only 300x300 input using a single Nvidia 1080 GPU. Especially, for small objects, DDSSD achieves 10.5% on MS COCO and 22.8% on FLIR thermal dataset, outperforming a lot of state-of-the-art object detection algorithms in both aspects of accuracy and speed.

cs.CV↗

arXiv2021

Generating Superpixels for High-resolution Images with Decoupled Patch Calibration

Superpixel segmentation has recently seen important progress benefiting from the advances in differentiable deep learning. However, the very high-resolution superpixel segmentation still remains challenging due to the expensive memory and computation cost, making the current advanced superpixel networks fail to process. In this paper, we devise Patch Calibration Networks (PCNet), aiming to efficiently and accurately implement high-resolution superpixel segmentation. PCNet follows the principle of producing high-resolution output from low-resolution input for saving GPU memory and relieving computation cost. To recall the fine details destroyed by the down-sampling operation, we propose a novel Decoupled Patch Calibration (DPC) branch for collaboratively augment the main superpixel generation branch. In particular, DPC takes a local patch from the high-resolution images and dynamically generates a binary mask to impose the network to focus on region boundaries. By sharing the parameters of DPC and main branches, the fine-detailed knowledge learned from high-resolution patches will be transferred to help calibrate the destroyed information. To the best of our knowledge, we make the first attempt to consider the deep-learning-based superpixel generation for high-resolution cases. To facilitate this research, we build evaluation benchmarks from two public datasets and one new constructed one, covering a wide range of diversities from fine-grained human parts to cityscapes. Extensive experiments demonstrate that our PCNet can not only perform favorably against the state-of-the-arts in the quantitative results but also improve the resolution upper bound from 3K to 5K on 1080Ti GPUs.

cs.CV↗

arXiv2021

Stable Polar Oxynitrides through Epitaxial Strain

We investigate energetically favorable structures of ABO$_2$N oxynitrides as functions of pressure and strain via swarm-intelligence-based structure prediction methods, DFT lattice dynamics and first-principles molecular dynamics. We predict several thermodynamically stable polar oxynitride perovskites under high pressures. In addition, we find that ferroelectric polar phases of perovskite-structured oxynitrides can be thermodynamically stable and synthesized at high pressure on appropriate substrates. The dynamical stability of the ferroelectric oxynitrides under epitaxial strain at ambient pressure also imply the possibility to synthesize them using pulsed laser deposition or other atomic layer deposition methods. Our results have broad implications for further exploration of other oxynitride materials as well. We performed first-principles molecular dynamics and find that the polar perovskite of YSiO$_2$N is metastable up to at least 600 K under compressive epitaxial strain before converting to the stable wollastonite-like structures. YSiO$_2$N is stabilized under pressure with extensional epitaxial strain. We predict that LaSi$_2$N, LaGeO$_2$N, BiSiO$_2$N, and BiGeO$_2$N are metastable as ferroelectric perovskites at zero pressure even without epitaxial strain.

cond-mat.mtrl-sci↗

arXiv2021

1st Place Solution to ICDAR 2021 RRC-ICTEXT End-to-end Text Spotting and Aesthetic Assessment on Integrated Circuit

This paper presents our proposed methods to ICDAR 2021 Robust Reading Challenge - Integrated Circuit Text Spotting and Aesthetic Assessment (ICDAR RRC-ICTEXT 2021). For the text spotting task, we detect the characters on integrated circuit and classify them based on yolov5 detection model. We balance the lowercase and non-lowercase by using SynthText, generated data and data sampler. We adopt semi-supervised algorithm and distillation to furtherly improve the model's accuracy. For the aesthetic assessment task, we add a classification branch of 3 classes to differentiate the aesthetic classes of each character. Finally, we make model deployment to accelerate inference speed and reduce memory consumption based on NVIDIA Tensorrt. Our methods achieve 59.1 mAP on task 3.1 with 31 FPS and 306M memory (rank 1), 78.7\% F2 score on task 3.2 with 30 FPS and 306M memory (rank 1).

cs.CV↗

arXiv2021

Sketch-Guided Scenery Image Outpainting

The outpainting results produced by existing approaches are often too random to meet users' requirement. In this work, we take the image outpainting one step forward by allowing users to harvest personal custom outpainting results using sketches as the guidance. To this end, we propose an encoder-decoder based network to conduct sketch-guided outpainting, where two alignment modules are adopted to impose the generated content to be realistic and consistent with the provided sketches. First, we apply a holistic alignment module to make the synthesized part be similar to the real one from the global view. Second, we reversely produce the sketches from the synthesized part and encourage them be consistent with the ground-truth ones using a sketch alignment module. In this way, the learned generator will be imposed to pay more attention to fine details and be sensitive to the guiding sketches. To our knowledge, this work is the first attempt to explore the challenging yet meaningful conditional scenery image outpainting. We conduct extensive experiments on two collected benchmarks to qualitatively and quantitatively validate the effectiveness of our approach compared with the other state-of-the-art generative models.

cs.CV↗

arXiv2020

DONet: Dual Objective Networks for Skin Lesion Segmentation

Skin lesion segmentation is a crucial step in the computer-aided diagnosis of dermoscopic images. In the last few years, deep learning based semantic segmentation methods have significantly advanced the skin lesion segmentation results. However, the current performance is still unsatisfactory due to some challenging factors such as large variety of lesion scale and ambiguous difference between lesion region and background. In this paper, we propose a simple yet effective framework, named Dual Objective Networks (DONet), to improve the skin lesion segmentation. Our DONet adopts two symmetric decoders to produce different predictions for approaching different objectives. Concretely, the two objectives are actually defined by different loss functions. In this way, the two decoders are encouraged to produce differentiated probability maps to match different optimization targets, resulting in complementary predictions accordingly. The complementary information learned by these two objectives are further aggregated together to make the final prediction, by which the uncertainty existing in segmentation maps can be significantly alleviated. Besides, to address the challenge of large variety of lesion scales and shapes in dermoscopic images, we additionally propose a recurrent context encoding module (RCEM) to model the complex correlation among skin lesions, where the features with different scale contexts are efficiently integrated to form a more robust representation. Extensive experiments on two popular benchmarks well demonstrate the effectiveness of the proposed DONet. In particular, our DONet achieves 0.881 and 0.931 dice score on ISIC 2018 and $\text{PH}^2$, respectively. Code will be made public available.

eess.IV↗

arXiv2020

Prediction of an Extended Ferroelectric Clathrate

Using first-principles calculations, we predict a lightweight room-temperature ferroelectric carbon-boron framework in a host/guest clathrate structure. This ferroelectric clathrate, with composition ScB$_3$C$_3$, exhibits high polarization density and low mass density compared with widely used commercial ferroelectrics. Molecular dynamics simulations show spontaneous polarization with a moderate above-room-temperature T$_c$ of $\sim$370 K, which implies large susceptibility and possibly large electrocaloric and piezoelectric constants at room temperature. Our findings open the possibility for a new class of ferroelectric materials with potential across a broad range of applications.

cond-mat.mtrl-sci↗

arXiv2020

Localization-aware Channel Pruning for Object Detection

Channel pruning is one of the important methods for deep model compression. Most of existing pruning methods mainly focus on classification. Few of them conduct systematic research on object detection. However, object detection is different from classification, which requires not only semantic information but also localization information. In this paper, based on discrimination-aware channel pruning (DCP) which is state-of-the-art pruning method for classification, we propose a localization-aware auxiliary network to find out the channels with key information for classification and regression so that we can conduct channel pruning directly for object detection, which saves lots of time and computing resources. In order to capture the localization information, we first design the auxiliary network with a contextual ROIAlign layer which can obtain precise localization information of the default boxes by pixel alignment and enlarges the receptive fields of the default boxes when pruning shallow layers. Then, we construct a loss function for object detection task which tends to keep the channels that contain the key information for classification and regression. Extensive experiments demonstrate the effectiveness of our method. On MS COCO, we prune 70\% parameters of the SSD based on ResNet-50 with modest accuracy drop, which outperforms the-state-of-art method.

cs.CV↗

arXiv2019

IoU-uniform R-CNN: Breaking Through the Limitations of RPN

Region Proposal Network (RPN) is the cornerstone of two-stage object detectors, it generates a sparse set of object proposals and alleviates the extrem foregroundbackground class imbalance problem during training. However, we find that the potential of the detector has not been fully exploited due to the IoU distribution imbalance and inadequate quantity of the training samples generated by RPN. With the increasing intersection over union (IoU), the exponentially smaller numbers of positive samples would lead to the distribution skewed towards lower IoUs, which hinders the optimization of detector at high IoU levels. In this paper, to break through the limitations of RPN, we propose IoU-Uniform R-CNN, a simple but effective method that directly generates training samples with uniform IoU distribution for the regression branch as well as the IoU prediction branch. Besides, we improve the performance of IoU prediction branch by eliminating the feature offsets of RoIs at inference, which helps the NMS procedure by preserving accurately localized bounding box. Extensive experiments on the PASCAL VOC and MS COCO dataset show the effectiveness of our method, as well as its compatibility and adaptivity to many object detection architectures. The code is made publicly available at https://github.com/zl1994/IoU-Uniform-R-CNN,

cs.CV↗

arXiv2019

Realization of metallic state in 1T-TaS2 with persisting long-range order of charge density wave

Metallization of 1T-TaS2 is generally initiated at the domain boundary of charge density wave (CDW), at the expense of its long-range order. However, we demonstrate in this study that the metallization of 1T-TaS2 can be also realized without breaking the long-range CDW order upon surface alkali doping. By using scanning tunneling microscopy, we find the long-range CDW order is always persisting, and the metallization is instead associated with additional in-gap excitations. Interestingly, the in-gap excitation is near the top of the lower Hubbard band, in contrast to a conventional electron-doped Mott insulator where it is beneath the upper Hubbard band. In combination with the numerical calculations, we suggest that the appearance of the in-gap excitations near the lower Hubbard band is mainly due to the effectively reduced on-site Coulomb energy by the adsorbed alkali ions.

cond-mat.str-el↗

arXiv2019

VrR-VG: Refocusing Visually-Relevant Relationships

Relationships encode the interactions among individual instances, and play a critical role in deep visual scene understanding. Suffering from the high predictability with non-visual information, existing methods tend to fit the statistical bias rather than ``learning'' to ``infer'' the relationships from images. To encourage further development in visual relationships, we propose a novel method to automatically mine more valuable relationships by pruning visually-irrelevant ones. We construct a new scene-graph dataset named Visually-Relevant Relationships Dataset (VrR-VG) based on Visual Genome. Compared with existing datasets, the performance gap between learnable and statistical method is more significant in VrR-VG, and frequency-based analysis does not work anymore. Moreover, we propose to learn a relationship-aware representation by jointly considering instances, attributes and relationships. By applying the representation-aware feature learned on VrR-VG, the performances of image captioning and visual question answering are systematically improved with a large margin, which demonstrates the gain of our dataset and the features embedding schema. VrR-VG is available via http://vrr-vg.com/.

cs.CV↗

arXiv2019

Phase Transition Pathway Sampling via Swarm Intelligence and Graph Theory

The prediction of reaction pathways for solid-solid transformations remains a key challenge. Here, we develop a pathway sampling method via swarm intelligence and graph theory, and demonstrate that our PALLAS method is an effective tool to help understand phase transformations in solid-state systems. The method is capable of finding low-energy transition pathways between two minima without having to specify any details of transition mechanism a priori. We benchmarked our PALLAS method against known phase transitions in cadmium selenide (CdSe) and silicon (Si). PALLAS readily identifies previously-reported, low-energy phase transition pathways for the wurtzite to rock-salt transition in CdSe and reveals a novel lower-energy pathway that has not yet been observed. In addition, PALLAS provides detailed information that explains the complex phase transition sequence observed during the decompression of Si from high pressure. Given the efficiency to identify low-barrier-energy reaction pathways, the PALLAS methodology represents a promising tool for materials by design with valuable insights for novel synthesis.

cond-mat.mtrl-sci↗

arXiv2019

Smart systems, the fourth industrial revolution and new challenges in distributed computing

Smart systems and the smart world concept are addressed in the framework of the fourth industrial revolution. New challenges in distributed autonomous robots and computing are considered. An illustration of a new kind of smart and reconfigurable distributed modular robot system is given. A prototype is also presented as well as the associated distributed algorithm.

cs.DC↗

arXiv2019

A programmable actuator for combined motion and connection and its application to modular robot

This paper proposes a new type of actuator at millimeter scale, which is based on Simplified Electro-Permanent (SEP) magnets. The new actuator can achieve connection and smooth motion by controlling the polarity of SEP magnets. Analyses based on numerical simulation are used to design a prototype. A dead-time controllable H-bridge and its multiplex design are proposed for controlling the new actuator and simplifying the electronic circuit. Finally, the new actuator is implemented in the DILI modular reconfigurable robot system. The experimental results show that with this new actuator, the DILI module can move smoothly and connect to other modules without power supply during connection. The maximum speed of DILI module is 20mm/s.

cs.RO↗

arXiv2019

Turning ZrTe5 into semiconductor through atomic intercalation

In this work, we use the liquid ammonia method to successfully intercalate potassium atoms into ZrTe5 single crystal, and find a transition from semimetal to semiconductor at low temperature in the intercalated ZrTe5. The resistance anomalous peak is gradually suppressed and finally disappears with increasing potassium concentration. Whilst, the according sign reversal is always observed in the Hall resistance measurement. We tentatively attribute the semimetal-semiconductor transition to the lattice expansion induced by atomic intercalation and thereby a larger energy band gap.

cond-mat.mtrl-sci↗

arXiv2019

Superconductivity in Potassium-intercalated Td-WTe2

To realize topological superconductor is one of the most attracting topics because of its great potential in quantum computation. In this study, we successfully intercalate potassium (K) into the van der Waals gap of type II Weyl semimetal WTe2, and discover the superconducting state in KxWTe2 through both electrical transport and scanning tunneling spectroscopy measurements. The superconductivity exhibits an evident anisotropic behavior. Moreover, we also uncover the coexistence of superconductivity and the positive magneto-resistance state. Structural analysis substantiates the negligible lattice expansion induced by the intercalation, therefore suggesting K-intercalated WTe2 still hosts the topological nontrivial state. These results indicate that the K-intercalated WTe2 may be a promising candidate to explore the topological superconductor.

cond-mat.supr-con↗

arXiv2018

Multi-view Point Cloud Registration with Adaptive Convergence Threshold and its Application on 3D Model Retrieval

Multi-view point cloud registration is a hot topic in the communities of multimedia technology and artificial intelligence (AI). In this paper, we propose a framework to reconstruct the 3D models by the multi-view point cloud registration algorithm with adaptive convergence threshold, and subsequently apply it to 3D model retrieval. The iterative closest point (ICP) algorithm is implemented combining with the motion average algorithm for the registration of multi-view point clouds. After the registration process, we design applications for 3D model retrieval. The geometric saliency map is computed based on the vertex curvature. The test facial triangle is then generated based on the saliency map, which is applied to compare with the standard facial triangle. The face and non-face models are then discriminated. The experiments and comparisons prove the effectiveness of the proposed framework.

cs.CV↗