arXiv subjects

Qin Liu

Publications and source records attributed to Qin Liu.

At least 73 records · Page 4Linked to original sources

arXiv2023

Deep Learning Methods for Real-time Detection and Analysis of Wagner Ulcer Classification System

At present, the ubiquity method to diagnose the severity of diabetic feet (DF) depends on professional podiatrists. However, in most cases, professional podiatrists have a heavy workload, especially in underdeveloped and developing countries and regions, and there are often insufficient podiatrists to meet the rapidly growing treatment needs of DF patients. It is necessary to develop a medical system that assists in diagnosing DF in order to reduce part of the workload for podiatrists and to provide timely relevant information to patients with DF. In this paper, we have developed a system that can classify and locate Wagner ulcers of diabetic foot in real-time. First, we proposed a dataset of 2688 diabetic feet with annotations. Then, in order to enable the system to detect diabetic foot ulcers in real time and accurately, this paper is based on the YOLOv3 algorithm coupled with image fusion, label smoothing, and variant learning rate mode technologies to improve the robustness and predictive accuracy of the original algorithm. Finally, the refinements on YOLOv3 was used as the optimal algorithm in this paper to deploy into Android smartphone to predict the classes and localization of the diabetic foot with real-time. The experimental results validate that the improved YOLOv3 algorithm achieves a mAP of 91.95%, and meets the needs of real-time detection and analysis of diabetic foot Wagner Ulcer on mobile devices, such as smart phones. This work has the potential to lead to a paradigm shift for clinical treatment of the DF in the future, to provide an effective healthcare solution for DF tissue analysis and healing status.

cs.CV↗

arXiv2023

SimpleClick: Interactive Image Segmentation with Simple Vision Transformers

Click-based interactive image segmentation aims at extracting objects with a limited user clicking. A hierarchical backbone is the de-facto architecture for current methods. Recently, the plain, non-hierarchical Vision Transformer (ViT) has emerged as a competitive backbone for dense prediction tasks. This design allows the original ViT to be a foundation model that can be finetuned for downstream tasks without redesigning a hierarchical backbone for pretraining. Although this design is simple and has been proven effective, it has not yet been explored for interactive image segmentation. To fill this gap, we propose SimpleClick, the first interactive segmentation method that leverages a plain backbone. Based on the plain backbone, we introduce a symmetric patch embedding layer that encodes clicks into the backbone with minor modifications to the backbone itself. With the plain backbone pretrained as a masked autoencoder (MAE), SimpleClick achieves state-of-the-art performance. Remarkably, our method achieves 4.15 NoC@90 on SBD, improving 21.8% over the previous best result. Extensive evaluation on medical images demonstrates the generalizability of our method. We further develop an extremely tiny ViT backbone for SimpleClick and provide a detailed computational analysis, highlighting its suitability as a practical annotation tool.

cs.CV↗

arXiv2022

PseudoClick: Interactive Image Segmentation with Click Imitation

The goal of click-based interactive image segmentation is to obtain precise object segmentation masks with limited user interaction, i.e., by a minimal number of user clicks. Existing methods require users to provide all the clicks: by first inspecting the segmentation mask and then providing points on mislabeled regions, iteratively. We ask the question: can our model directly predict where to click, so as to further reduce the user interaction cost? To this end, we propose {\PseudoClick}, a generic framework that enables existing segmentation networks to propose candidate next clicks. These automatically generated clicks, termed pseudo clicks in this work, serve as an imitation of human clicks to refine the segmentation mask.

cs.CV↗

arXiv2022

iSegFormer: Interactive Segmentation via Transformers with Application to 3D Knee MR Images

We propose iSegFormer, a memory-efficient transformer that combines a Swin transformer with a lightweight multilayer perceptron (MLP) decoder. With the efficient Swin transformer blocks for hierarchical self-attention and the simple MLP decoder for aggregating both local and global attention, iSegFormer learns powerful representations while achieving high computational efficiencies. Specifically, we apply iSegFormer to interactive 3D medical image segmentation.

cs.CV↗

arXiv2022

Statistical Dependency Guided Contrastive Learning for Multiple Labeling in Prenatal Ultrasound

Standard plane recognition plays an important role in prenatal ultrasound (US) screening. Automatically recognizing the standard plane along with the corresponding anatomical structures in US image can not only facilitate US image interpretation but also improve diagnostic efficiency. In this study, we build a novel multi-label learning (MLL) scheme to identify multiple standard planes and corresponding anatomical structures of fetus simultaneously. Our contribution is three-fold. First, we represent the class correlation by word embeddings to capture the fine-grained semantic and latent statistical concurrency. Second, we equip the MLL with a graph convolutional network to explore the inner and outer relationship among categories. Third, we propose a novel cluster relabel-based contrastive learning algorithm to encourage the divergence among ambiguous classes. Extensive validation was performed on our large in-house dataset. Our approach reports the highest accuracy as 90.25% for standard planes labeling, 85.59% for planes and structures labeling and mAP as 94.63%. The proposed MLL scheme provides a novel perspective for standard plane recognition and can be easily extended to other medical image classification tasks.

cs.CV↗

arXiv2021

SkullEngine: A Multi-stage CNN Framework for Collaborative CBCT Image Segmentation and Landmark Detection

We propose a multi-stage coarse-to-fine CNN-based framework, called SkullEngine, for high-resolution segmentation and large-scale landmark detection through a collaborative, integrated, and scalable JSD model and three segmentation and landmark detection refinement models. We evaluated our framework on a clinical dataset consisting of 170 CBCT/CT images for the task of segmenting 2 bones (midface and mandible) and detecting 175 clinically common landmarks on bones, teeth, and soft tissues.

eess.IV↗

arXiv2021

A Self-Supervised Deep Framework for Reference Bony Shape Estimation in Orthognathic Surgical Planning

Virtual orthognathic surgical planning involves simulating surgical corrections of jaw deformities on 3D facial bony shape models. Due to the lack of necessary guidance, the planning procedure is highly experience-dependent and the planning results are often suboptimal. A reference facial bony shape model representing normal anatomies can provide an objective guidance to improve planning accuracy. Therefore, we propose a self-supervised deep framework to automatically estimate reference facial bony shape models. Our framework is an end-to-end trainable network, consisting of a simulator and a corrector. In the training stage, the simulator maps jaw deformities of a patient bone to a normal bone to generate a simulated deformed bone. The corrector then restores the simulated deformed bone back to normal. In the inference stage, the trained corrector is applied to generate a patient-specific normal-looking reference bone from a real deformed bone. The proposed framework was evaluated using a clinical dataset and compared with a state-of-the-art method that is based on a supervised point-cloud network. Experimental results show that the estimated shape models given by our approach are clinically acceptable and significantly more accurate than that of the competing method.

cs.CV↗

arXiv2021

Towards Robust Active Feature Acquisition

Truly intelligent systems are expected to make critical decisions with incomplete and uncertain data. Active feature acquisition (AFA), where features are sequentially acquired to improve the prediction, is a step towards this goal. However, current AFA models all deal with a small set of candidate features and have difficulty scaling to a large feature space. Moreover, they are ignorant about the valid domains where they can predict confidently, thus they can be vulnerable to out-of-distribution (OOD) inputs. In order to remedy these deficiencies and bring AFA models closer to practical use, we propose several techniques to advance the current AFA approaches. Our framework can easily handle a large number of features using a hierarchical acquisition policy and is more robust to OOD inputs with the help of an OOD detector for partially observed data. Extensive experiments demonstrate the efficacy of our framework over strong baselines.

cs.LG↗

arXiv2021

TextFlint: Unified Multilingual Robustness Evaluation Toolkit for Natural Language Processing

Various robustness evaluation methodologies from different perspectives have been proposed for different natural language processing (NLP) tasks. These methods have often focused on either universal or task-specific generalization capabilities. In this work, we propose a multilingual robustness evaluation platform for NLP tasks (TextFlint) that incorporates universal text transformation, task-specific transformation, adversarial attack, subpopulation, and their combinations to provide comprehensive robustness analysis. TextFlint enables practitioners to automatically evaluate their models from all aspects or to customize their evaluations as desired with just a few lines of code. To guarantee user acceptability, all the text transformations are linguistically based, and we provide a human evaluation for each one. TextFlint generates complete analytical reports as well as targeted augmented data to address the shortcomings of the model's robustness. To validate TextFlint's utility, we performed large-scale empirical evaluations (over 67,000 evaluations) on state-of-the-art deep learning models, classic supervised methods, and real-world systems. Almost all models showed significant performance degradation, including a decline of more than 50% of BERT's prediction accuracy on tasks such as aspect-level sentiment classification, named entity recognition, and natural language inference. Therefore, we call for the robustness to be included in the model evaluation, so as to promote the healthy development of NLP technology.

cs.CL↗

arXiv2020

Kondo scenario of the γ-α phase transition in single crystalline Cerium thin films

The physical mechanism driving the $γ$-$α$ phase transition of face-centre-cubic (fcc) cerium (Ce) remains controversial until now. In this work, high quality single crystalline fcc-Ce thin films were grown on Graphene/6$H$-SiC(0001) substrate, and explored by XRD and ARPES measurement. XRD spectra showed a clear $γ$-$α$ phase transition at $T_{γ-α}\approx$ 50 K, which is retarded by strain effect from substrate comparing with $T_{γ-α}$ (about 140 K) of the bulk Ce metal. However, APRES spectra did not show any signature of $α$-phase emerging in the surface-layer from 300 K to 17 K, which implied that $α$-phase might form at the bulk-layer of our Ce thin films. Besides, an evident Kondo dip near Fermi energy was observed in the APRES spectrum at 80 K, indicting the formation of Kondo singlet states in $γ$-Ce. Furthermore, the DFT+DMFT calculations were performed to simulate the electronic structures and the theoretical spectral functions agreed well with the experimental ARPES spectra. In $γ$-Ce, the behavior of the self-energy's imaginary part at low frequency not only confirmed that the Kondo singlet states emerged at $T_{\rm KS} \geq 80$ K, but also implied that they became coherent states at a lower characteristic temperature ($T_{\rm coh}\sim 40$ K) due to the indirect RKKY interaction among $f$-$f$ electrons. Besides, $T_{\rm coh}$ from the theoretical simulation was close to $T_{γ-α}$ from the XRD spectra. These issues suggested that the Kondo scenario might play an important role in the $γ$-$α$ phase transition of cerium thin films.

cond-mat.str-el↗

arXiv2020

Observation of discrete conventional Caroli-de Gennes-Matricon states in the vortex core of single-layer FeSe/SrTiO3

Using low-temperature scanning tunneling microscopy (STM), we studied the vortex states of single-layer FeSe film on SrTiO3 (100) substrate, and the local behaviors of superconductivity at sample boundaries. We clearly observed multiple discrete Caroli-de Gennes-Matricon (CdGM) states in the vortex core, and quantitative analysis shows their energies well follow the formula: E = μΔ^2/E_F, where μ is a half integer and Δ is the mean superconducting gap over the Fermi surface. Meanwhile, a fully gapped spectrum without states near zero bias is observed at [110](Fe) oriented boundary of 1 ML and 2 ML FeSe films, and atomic step edge of 1 ML FeSe. Accompanied with theoretical calculations, our results indicate a s-wave pairing without sign-change in the high-TC FeSe_SrTiO3 superconductor.

cond-mat.supr-con↗

arXiv2019

Exploring the exotic f states of prototype compounds CeSb and USb

To unravel the interplay between the strong electronic correlation and itinerant-localized dual nature in atypical f electron systems, we employed the density functional theory in combination with the single-site dynamical mean-field theory to systematically investigate the electronic structures of CeSb and USb. We find that the 4f states in CeSb are mostly localized which show a weak quasi-particle resonance peak near the Fermi level. Conversely, the 5f electrons in USb display partially itinerant feature, accompanied by mixed-valence behavior and prominent valence state fluctuations. Particularly, the 4f electronic correlations in CeSb are distinctly orbital-selective with strikingly renormalized 4f5/2 states, according to the low-energy behaviors of 4f self-energy functions. It is believed that the strong electronic correlation and fantastic bonding of f states contribute to elucidate the fascinating magnetism.

cond-mat.str-el↗

arXiv2019

Improving Model Drift for Robust Object Tracking

Discriminative correlation filters show excellent performance in object tracking. However, in complex scenes, the apparent characteristics of the tracked target are variable, which makes it easy to pollute the model and cause the model drift. In this paper, considering that the secondary peak has a greater impact on the model update, we propose a method for detecting the primary and secondary peaks of the response map. Secondly, a novel confidence function which uses the adaptive update discriminant mechanism is proposed, which yield good robustness. Thirdly, we propose a robust tracker with correlation filters, which uses hand-crafted features and can improve model drift in complex scenes. Finally, in order to cope with the current trackers' multi-feature response merge, we propose a simple exponential adaptive merge approach. Extensive experiments are performed on OTB2013, OTB100 and TC128 datasets. Our approach performs superiorly against several state-of-the-art trackers while runs at speed in real time.

cs.CV↗

arXiv2018

A vertex reconstruction algorithm in the central detector of JUNO

The Jiangmen Underground Neutrino Observatory (JUNO) is designed to study neutrino mass hierarchy and measure three of the neutrino oscillation parameters with high precision using reactor antineutrinos. It is also able to study many other physical phenomena, including supernova neutrinos, solar neutrinos, geo-neutrinos, atmosphere neutrinos, and so forth. The central detector of JUNO contains 20,000~tons of liquid scintillator (LS) and about 18,000 20-inch photomultiplier tubes (PMTs), which is the largest liquid scintillator one under construction in the world up today. The energy resolution is expected to be 3\%/$\sqrt{E(MeV)}$. To meet the requirements of the experiment, an algorithm of vertex reconstruction, which takes into account time and charge information of PMTs, has been developed by deploying the maximum likelihood method and well understanding the complicated optical processes in the liquid scintillator.

physics.ins-det↗

arXiv2018

Efficiency and power of minimally nonlinear irreversible heat engines with broken time-reversal symmetry

We study the minimally nonlinear irreversible heat engines in which the time-reversal symmetry for the systems may b e broken. The expressions for the power and the efficiency are derived, in which the effects of the nonlinear terms due to dissipations are included. We show that, as within the linear responses, the minimally nonlinear irreversible heat engines enable attainment of Carnot efficiency at positive power. We also find that the Curzon-Ahlborn limit imposed on the efficiency at maximum power can be overcomed if the time-reversal symmetry is broken.

cond-mat.stat-mech↗

arXiv2016

Three-dimensional bulk electronic structure of the Kondo lattice CeIn3 revealed by photoemission

We show the three-dimensional electronic structure of the Kondo lattice CeIn3 using soft x-ray angle resolved photoemission spectroscopy in the paramagnetic state. For the first time, we have directly observed the three-dimensional topology of the Fermi surface of CeIn3 by photoemission. The Fermi surface has a complicated hole pocket centred at the Γ-Z line and an elliptical electron pocket centred at the R point of the Brillouin zone. Polarization and photon-energy dependent photoemission results both indicate the nearly localized nature of the 4f electrons in CeIn3, consistent with the theoretical prediction by means of the combination of density functional theory and single-site dynamical meanfield theory. Those results illustrate that the f electrons of CeIn3, which is the parent material of CeMIn5 compounds, are closer to the localized description than the layered CeMIn5 compounds.

cond-mat.str-el↗

arXiv2016

PowerWalk: Scalable Personalized PageRank via Random Walks with Vertex-Centric Decomposition

Most methods for Personalized PageRank (PPR) precompute and store all accurate PPR vectors, and at query time, return the ones of interest directly. However, the storage and computation of all accurate PPR vectors can be prohibitive for large graphs, especially in caching them in memory for real-time online querying. In this paper, we propose a distributed framework that strikes a better balance between offline indexing and online querying. The offline indexing attains a fingerprint of the PPR vector of each vertex by performing billions of "short" random walks in parallel across a cluster of machines. We prove that our indexing method has an exponential convergence, achieving the same precision with previous methods using a much smaller number of random walks. At query time, the new PPR vector is composed by a linear combination of related fingerprints, in a highly efficient vertex-centric decomposition manner. Interestingly, the resulting PPR vector is much more accurate than its offline counterpart because it actually uses more random walks in its estimation. More importantly, we show that such decomposition for a batch of queries can be very efficiently processed using a shared decomposition. Our implementation, PowerWalk, takes advantage of advanced distributed graph engines and it outperforms the state-of-the-art algorithms by orders of magnitude. Particularly, it responses to tens of thousands of queries on graphs with billions of edges in just a few seconds.

cs.IR↗

arXiv2016

Time and Frequency Injection into a Stabilized Fiber Link for Multi-clock Dissemination Network

Owing to the characteristics of ultra-low loss and anti-electromagnetic interference, using optical fiber to deliver time and frequency signal has been a preferred choice for high precise clock dissemination and comparison. As a brilliant idea, one has been able to reproduce ultra-stable signals from one local station to multiple users. In this paper, we take a step further. A concept of multi-clock (in different locations) dissemination for multi-terminals is presented. By injecting frequency signals into one stabilized ring-like fiber network, the relative stabilities of 3.4e-14@1s for a master clock dissemination and 5.1e-14@1s for a slave clock dissemination have been achieved. The proposed scheme can greatly simplify the future 'N' to 'N' time and frequency dissemination network, especially facing a multi-clock comparison situation.

physics.optics↗