arXiv subjects

Hongliang Li

Publications and source records attributed to Hongliang Li.

95 records · Page 6Linked to original sources

arXiv2019

Hierarchy Neighborhood Discriminative Hashing for An Unified View of Single-Label and Multi-Label Image retrieval

Recently, deep supervised hashing methods have become popular for large-scale image retrieval task. To preserve the semantic similarity notion between examples, they typically utilize the pairwise supervision or the triplet supervised information for hash learning. However, these methods usually ignore the semantic class information which can help the improvement of the semantic discriminative ability of hash codes. In this paper, we propose a novel hierarchy neighborhood discriminative hashing method. Specifically, we construct a bipartite graph to build coarse semantic neighbourhood relationship between the sub-class feature centers and the embeddings features. Moreover, we utilize the pairwise supervised information to construct the fined semantic neighbourhood relationship between embeddings features. Finally, we propose a hierarchy neighborhood discriminative hashing loss to unify the single-label and multilabel image retrieval problem with a one-stream deep neural network architecture. Experimental results on two largescale datasets demonstrate that the proposed method can outperform the state-of-the-art hashing methods.

cs.CV↗

arXiv2018

New Nonconforming Elements for Linear Strain Gradient Elastic Model

Based on a new H$^2-$Korn's inequality, we propose new nonconforming elements for the linear strain gradient elastic model. The first group of elements are H$^1-$conforming but H$^2-$nonconforming. The tensor product NTW element [Tai:2001] and the tensor product Specht triangle are two typical representatives. The second element is based on Morley's triangle with a modified elastic strain energy. We proved new interpolation error estimates for all these elements, which are key to prove uniform rates of convergence for the proposed elements. Numerical results are reported and they are consistent with the theoretical prediction.

math.NA↗

arXiv2017

A Perceptually Weighted Rank Correlation Indicator for Objective Image Quality Assessment

In the field of objective image quality assessment (IQA), the Spearman's $ρ$ and Kendall's $τ$ are two most popular rank correlation indicators, which straightforwardly assign uniform weight to all quality levels and assume each pair of images are sortable. They are successful for measuring the average accuracy of an IQA metric in ranking multiple processed images. However, two important perceptual properties are ignored by them as well. Firstly, the sorting accuracy (SA) of high quality images are usually more important than the poor quality ones in many real world applications, where only the top-ranked images would be pushed to the users. Secondly, due to the subjective uncertainty in making judgement, two perceptually similar images are usually hardly sortable, whose ranks do not contribute to the evaluation of an IQA metric. To more accurately compare different IQA algorithms, we explore a perceptually weighted rank correlation indicator in this paper, which rewards the capability of correctly ranking high quality images, and suppresses the attention towards insensitive rank mistakes. More specifically, we focus on activating `valid' pairwise comparison towards image quality, whose difference exceeds a given sensory threshold (ST). Meanwhile, each image pair is assigned an unique weight, which is determined by both the quality level and rank deviation. By modifying the perception threshold, we can illustrate the sorting accuracy with a more sophisticated SA-ST curve, rather than a single rank correlation coefficient. The proposed indicator offers a new insight for interpreting visual perception behaviors. Furthermore, the applicability of our indicator is validated in recommending robust IQA metrics for both the degraded and enhanced image data.

cs.CV↗

arXiv2017

Two robust nonconforming H$^2-$elements for linear strain gradient elasticity

We propose two nonconforming finite elements to approximate a boundary value problem arising from strain gradient elasticity, which is a higher-order perturbation of the linearized elastic system. Our elements are H$^2-$nonconforming while H$^1-$conforming. We show both elements converges in the energy norm uniformly with respect to the perturbation parameter.

math.NA↗

arXiv2015

Fast Transfer-free Synthesis of High-quality Monolayer Graphene on Insulating Substrates by Simple Rapid Thermal Treatment

The transfer-free synthesis of high-quality, large-area graphene on a given dielectric substrate, which is highly desirable for device applications, remains a significant challenge. In this paper, we report on a simple rapid thermal treatment method for the fast and direct growth of high-quality, large-scale monolayer graphene on a SiO2/Si substrate from solid carbon sources. The stack structure of solid carbon layer/copper film/SiO2 is adopted in the RTT process. The inserted copper film does not only act as an active catalyst for the carbon precursor but also serves as a "filter" that prevents premature carbon dissolution, and thus, contributes to monolayer graphene growth on SiO2/Si. The produced monolayer graphene exhibits high carrier mobility of up to 3000 cm2 V-1s-1 at room temperature and standard half-integer quantum oscillations.

cond-mat.mtrl-sci↗