arXiv subjects

Xu Wang

Publications and source records attributed to Xu Wang.

At least 343 records · Page 19Linked to original sources

arXiv2022

Expansion and Shrinkage of Localization for Weakly-Supervised Semantic Segmentation

Generating precise class-aware pseudo ground-truths, a.k.a, class activation maps (CAMs), is essential for weakly-supervised semantic segmentation. The original CAM method usually produces incomplete and inaccurate localization maps. To tackle with this issue, this paper proposes an Expansion and Shrinkage scheme based on the offset learning in the deformable convolution, to sequentially improve the recall and precision of the located object in the two respective stages. In the Expansion stage, an offset learning branch in a deformable convolution layer, referred as "expansion sampler" seeks for sampling increasingly less discriminative object regions, driven by an inverse supervision signal that maximizes image-level classification loss. The located more complete object in the Expansion stage is then gradually narrowed down to the final object region during the Shrinkage stage. In the Shrinkage stage, the offset learning branch of another deformable convolution layer, referred as "shrinkage sampler", is introduced to exclude the false positive background regions attended in the Expansion stage to improve the precision of the localization maps. We conduct various experiments on PASCAL VOC 2012 and MS COCO 2014 to well demonstrate the superiority of our method over other state-of-the-art methods for weakly-supervised semantic segmentation. Code will be made publicly available here https://github.com/TyroneLi/ESOL_WSSS.

cs.CV↗

arXiv2022

Semi-automatic Data Annotation System for Multi-Target Multi-Camera Vehicle Tracking

Multi-target multi-camera tracking (MTMCT) plays an important role in intelligent video analysis, surveillance video retrieval, and other application scenarios. Nowadays, the deep-learning-based MTMCT has been the mainstream and has achieved fascinating improvements regarding tracking accuracy and efficiency. However, according to our investigation, the lacking of datasets focusing on real-world application scenarios limits the further improvements for current learning-based MTMCT models. Specifically, the learning-based MTMCT models training by common datasets usually cannot achieve satisfactory results in real-world application scenarios. Motivated by this, this paper presents a semi-automatic data annotation system to facilitate the real-world MTMCT dataset establishment. The proposed system first employs a deep-learning-based single-camera trajectory generation method to automatically extract trajectories from surveillance videos. Subsequently, the system provides a recommendation list in the following manual cross-camera trajectory matching process. The recommendation list is generated based on side information, including camera location, timestamp relation, and background scene. In the experimental stage, extensive results further demonstrate the efficiency of the proposed system.

cs.CV↗

arXiv2022

Weakly Supervised Semantic Segmentation via Progressive Patch Learning

Most of the existing semantic segmentation approaches with image-level class labels as supervision, highly rely on the initial class activation map (CAM) generated from the standard classification network. In this paper, a novel "Progressive Patch Learning" approach is proposed to improve the local details extraction of the classification, producing the CAM better covering the whole object rather than only the most discriminative regions as in CAMs obtained in conventional classification models. "Patch Learning" destructs the feature maps into patches and independently processes each local patch in parallel before the final aggregation. Such a mechanism enforces the network to find weak information from the scattered discriminative local parts, achieving enhanced local details sensitivity. "Progressive Patch Learning" further extends the feature destruction and patch learning to multi-level granularities in a progressive manner. Cooperating with a multi-stage optimization strategy, such a "Progressive Patch Learning" mechanism implicitly provides the model with the feature extraction ability across different locality-granularities. As an alternative to the implicit multi-granularity progressive fusion approach, we additionally propose an explicit method to simultaneously fuse features from different granularities in a single model, further enhancing the CAM quality on the full object coverage. Our proposed method achieves outstanding performance on the PASCAL VOC 2012 dataset e.g., with 69.6$% mIoU on the test set), which surpasses most existing weakly supervised semantic segmentation methods. Code will be made publicly available here https://github.com/TyroneLi/PPL_WSSS.

cs.CV↗

arXiv2022

Curvature of the base manifold of a Monge-Ampère fibration and its existence

In this paper, we consider a special relative Kähler fibration that satisfies a homogenous Monge-Ampère equation, which is called a Monge-Ampère fibration. There exist two canonical types of generalized Weil-Petersson metrics on the base complex manifold of the fibration. For the second generalized Weil-Petersson metric, we obtain an explicit curvature formula and prove that the holomorphic bisectional curvature is non-positive, the holomorphic sectional curvature, the Ricci curvature, and the scalar curvature are all bounded from above by a negative constant. For a holomorphic vector bundle over a compact Kähler manifold, we prove that it admits a projectively flat Hermitian structure if and only if the associated projective bundle fibration is a Monge-Ampère fibration. In general, we can prove that a relative Kähler fibration is Monge-Ampère if and only if an associated infinite rank Higgs bundle is Higgs-flat. We also discuss some typical examples of Monge-Ampère fibrations.

math.AG↗

arXiv2022

Gaze Estimation Approach Using Deep Differential Residual Network

Gaze estimation, which is a method to determine where a person is looking at given the person's full face, is a valuable clue for understanding human intention. Similarly to other domains of computer vision, deep learning (DL) methods have gained recognition in the gaze estimation domain. However, there are still gaze calibration problems in the gaze estimation domain, thus preventing existing methods from further improving the performances. An effective solution is to directly predict the difference information of two human eyes, such as the differential network (Diff-Nn). However, this solution results in a loss of accuracy when using only one inference image. We propose a differential residual model (DRNet) combined with a new loss function to make use of the difference information of two eye images. We treat the difference information as auxiliary information. We assess the proposed model (DRNet) mainly using two public datasets (1) MpiiGaze and (2) Eyediap. Considering only the eye features, DRNet outperforms the state-of-the-art gaze estimation methods with $angular-error$ of 4.57 and 6.14 using MpiiGaze and Eyediap datasets, respectively. Furthermore, the experimental results also demonstrate that DRNet is extremely robust to noise images.

cs.CV↗

arXiv2022

Predictive and Contrastive: Dual-Auxiliary Learning for Recommendation

Self-supervised learning (SSL) recently has achieved outstanding success on recommendation. By setting up an auxiliary task (either predictive or contrastive), SSL can discover supervisory signals from the raw data without human annotation, which greatly mitigates the problem of sparse user-item interactions. However, most SSL-based recommendation models rely on general-purpose auxiliary tasks, e.g., maximizing correspondence between node representations learned from the original and perturbed interaction graphs, which are explicitly irrelevant to the recommendation task. Accordingly, the rich semantics reflected by social relationships and item categories, which lie in the recommendation data-based heterogeneous graphs, are not fully exploited. To explore recommendation-specific auxiliary tasks, we first quantitatively analyze the heterogeneous interaction data and find a strong positive correlation between the interactions and the number of user-item paths induced by meta-paths. Based on the finding, we design two auxiliary tasks that are tightly coupled with the target task (one is predictive and the other one is contrastive) towards connecting recommendation with the self-supervision signals hiding in the positive correlation. Finally, a model-agnostic DUal-Auxiliary Learning (DUAL) framework which unifies the SSL and recommendation tasks is developed. The extensive experiments conducted on three real-world datasets demonstrate that DUAL can significantly improve recommendation, reaching the state-of-the-art performance.

cs.IR↗

arXiv2022

Nuclear excitation cross section of $^{229}$Th via inelastic electron scattering

Nuclear excitation cross section of $^{229}$Th from the ground state to the low-lying isomeric state via inelastic electron scattering is calculated, on the level of Dirac distorted wave Born approximation. With electron energies below 100 eV, inelastic scattering is very efficient in the isomeric excitation, yielding excitation cross sections on the order of 10$^{-27}$ to 10$^{-26}$ cm$^2$. Systematic analyses are presented on elements affecting the excitation cross section, including the ion-core potential, the relativistic effect, the knowledge of the reduced nuclear transition probabilities, etc.

nucl-th↗

arXiv2022

Graph Property Prediction on Open Graph Benchmark: A Winning Solution by Graph Neural Architecture Search

Aiming at two molecular graph datasets and one protein association subgraph dataset in OGB graph classification task, we design a graph neural network framework for graph classification task by introducing PAS(Pooling Architecture Search). At the same time, we improve it based on the GNN topology design method F2GNN to further design the feature selection and fusion strategies, so as to further improve the performance of the model in the graph property prediction task while overcoming the over smoothing problem of deep GNN training. Finally, a performance breakthrough is achieved on these three datasets, which is significantly better than other methods with fixed aggregate function. It is proved that the NAS method has high generalization ability for multiple tasks and the advantage of our method in processing graph property prediction tasks.

cs.LG↗

arXiv2022

$λ$-domain VVC Rate Control Based on Game Theory

Versatile Video Coding (VVC) has set a new milestone in high-efficiency video coding. In the standard encoder, the $λ$-domain rate control is incorporated for its high accuracy and good Rate-Distortion (RD) performance. In this paper, we formulate this task as a Nash equilibrium problem that effectively bargains between multiple agents, {\it i.e.}, Coding Tree Units (CTUs) in the frame. After that, we calculate the optimal $λ$ value with a two-step strategy: a Newton method to iteratively obtain an intermediate variable, and a solution of Nash equilibrium to obtain the optimal $λ$. Finally, we propose an effective CTU-level rate allocation with the optimal $λ$ value. To the best of our knowledge, we are the first to combine game theory with $λ$-domain rate control. Experimental results with Common Test Conditions (CTC) demonstrate the efficiency of the proposed method, which outperforms the state-of-the-art CTU-level rate allocation algorithms.

cs.MM↗

arXiv2022

Recollision induced nuclear excitation of $^{229}$Th

Previously we proposed a new approach of exciting the $^{229}$Th nucleus using laser-driven electron recollision [W. Wang et al., Phys. Rev. Lett. 127, 052501 (2021)]. The current article is aimed to elaborate the method by explaining further theoretical details and presenting extended new results. The method has also been improved by adopting the electronic excitation cross sections calculated recently by Tkalya [E. V. Tkalya, Phys. Rev. Lett. 124, 242501 (2020)]. The new cross sections are obtained from Dirac distorted-wave calculations instead of from Dirac plane-wave calculations as we used previously. The distorted-wave cross sections are shown to be 5 to 6 orders of magnitude higher than the plane-wave results. With the excitation cross sections updated, the probability of isomeric excitation of $^{229}$Th from electron recollision is calculated to be on the order of $10^{-12}$ per nucleus per (femtosecond) laser pulse. Dependency of the excitation probability on various laser parameters is calculated and discussed, including the laser intensity, the laser wavelength, and the laser pulse duration.

physics.atom-ph↗

arXiv2022

Towards Process-Oriented, Modular, and Versatile Question Generation that Meets Educational Needs

NLP-powered automatic question generation (QG) techniques carry great pedagogical potential of saving educators' time and benefiting student learning. Yet, QG systems have not been widely adopted in classrooms to date. In this work, we aim to pinpoint key impediments and investigate how to improve the usability of automatic QG techniques for educational purposes by understanding how instructors construct questions and identifying touch points to enhance the underlying NLP models. We perform an in-depth need finding study with 11 instructors across 7 different universities, and summarize their thought processes and needs when creating questions. While instructors show great interests in using NLP systems to support question design, none of them has used such tools in practice. They resort to multiple sources of information, ranging from domain knowledge to students' misconceptions, all of which missing from today's QG systems. We argue that building effective human-NLP collaborative QG systems that emphasize instructor control and explainability is imperative for real-world adoption. We call for QG systems to provide process-oriented support, use modular design, and handle diverse sources of input.

cs.HC↗

arXiv2022

Emergence of superconducting dome in insulating ZrNx films via nitrogen manipulation

Reproducing the electronic phase diagram of strongly correlated high-transition-temperature (high-Tc) superconductors in materials other than Cu-, Fe-, and Ni-based compounds has been a challenging task. Only very recently, a few material systems have partially achieved this goal by band engineering. In this work, we combine film growth, charge transport, magnetometry, Terahertz Spectroscopy, Raman scattering, and Scanning Transmission Electron Microscopy to investigate superconductivity and the normal state of ZrNx, which reveals a phase diagram that bears extraordinary similarities to those of high-Tc superconductors. Remarkably, even though superconductivity of ZrNx can be characterized within the Bardeen-Cooper-Schrieffer paradigm and its normal state can be understood within the Fermi liquid framework, by tunning the N chemical concentration, we observe the evolution of a superconducting dome in the close vicinity of a strongly insulating state and a normal state resistivity mimics its counterpart of the high-Tc superconductors.

cond-mat.supr-con↗

arXiv2022

Periodic repeating fast radio bursts: interaction between a magnetized neutron star and its planet in an eccentric orbit

Fast radio bursts (FRBs) are mysterious transient phenomena. The study of repeating FRBs may provide useful information about their nature due to their redetectability. The two most famous repeating sources are FRBs 121102 and 180916, with a period of 157 days and 16.35 days, respectively. Previous studies suggest that the periodicity of FRBs is likely associated with neutron star (NS) binary systems. Here we introduce a new model which proposes that periodic repeating FRBs are due to the interaction of a NS with its planet in a highly elliptical orbit. The periastron of the planet is very close to the NS so that it would be partially disrupted by tidal force every time it passes through the periastron. Fragments generated in the process could interact with the compact star through the Alfvén wing mechanism and produce FRBs. The model can naturally explain the repeatability of FRBs with a period ranging from a few days to several hundred days, but it generally requires that the eccentricity of the planet's orbit should be large enough. Taking FRBs 121102 and 180916 as examples, it is shown that the main features of the observed repeating behaviors can be satisfactorily accounted for.

astro-ph.HE↗

arXiv2022

Boundary Corrected Multi-scale Fusion Network for Real-time Semantic Segmentation

Image semantic segmentation aims at the pixel-level classification of images, which has requirements for both accuracy and speed in practical application. Existing semantic segmentation methods mainly rely on the high-resolution input to achieve high accuracy and do not meet the requirements of inference time. Although some methods focus on high-speed scene parsing with lightweight architectures, they can not fully mine semantic features under low computation with relatively low performance. To realize the real-time and high-precision segmentation, we propose a new method named Boundary Corrected Multi-scale Fusion Network, which uses the designed Low-resolution Multi-scale Fusion Module to extract semantic information. Moreover, to deal with boundary errors caused by low-resolution feature map fusion, we further design an additional Boundary Corrected Loss to constrain overly smooth features. Extensive experiments show that our method achieves a state-of-the-art balance of accuracy and speed for the real-time semantic segmentation.

cs.CV↗

arXiv2022

Distortion-Aware Loop Filtering of Intra 360^o Video Coding with Equirectangular Projection

In this paper, we propose a distortion-aware loop filtering model to improve the performance of intra coding for 360$^o$ videos projected via equirectangular projection (ERP) format. To enable the awareness of distortion, our proposed module analyzes content characteristics based on a coding unit (CU) partition mask and processes them through partial convolution to activate the specified area. The feature recalibration module, which leverages cascaded residual channel-wise attention blocks (RCABs) to adjust the inter-channel and intra-channel features automatically, is capable of adapting with different quality levels. The perceptual geometry optimization combining with weighted mean squared error (WMSE) and the perceptual loss guarantees both the local field of view (FoV) and global image reconstruction with high quality. Extensive experimental results show that our proposed scheme achieves significant bitrate savings compared with the anchor (HM + 360Lib), leading to 8.9%, 9.0%, 7.1% and 7.4% on average bit rate reductions in terms of PSNR, WPSNR, and PSNR of two viewports for luminance component of 360^o videos, respectively.

cs.CV↗

arXiv2022

LighTN: Light-weight Transformer Network for Performance-overhead Tradeoff in Point Cloud Downsampling

Compared with traditional task-irrelevant downsampling methods, task-oriented neural networks have shown improved performance in point cloud downsampling range. Recently, Transformer family of networks has shown a more powerful learning capacity in visual tasks. However, Transformer-based architectures potentially consume too many resources which are usually worthless for low overhead task networks in downsampling range. This paper proposes a novel light-weight Transformer network (LighTN) for task-oriented point cloud downsampling, as an end-to-end and plug-and-play solution. In LighTN, a single-head self-correlation module is presented to extract refined global contextual features, where three projection matrices are simultaneously eliminated to save resource overhead, and the output of symmetric matrix satisfies the permutation invariant. Then, we design a novel downsampling loss function to guide LighTN focuses on critical point cloud regions with more uniform distribution and prominent points coverage. Furthermore, We introduce a feed-forward network scaling mechanism to enhance the learnable capacity of LighTN according to the expand-reduce strategy. The result of extensive experiments on classification and registration tasks demonstrates LighTN can achieve state-of-the-art performance with limited resource overhead.

cs.CV↗

arXiv2022

Real-space Observation of Incommensurate Spin Density Wave and Coexisting Charge Density Wave on Cr(001) surface

In itinerant magnetic systems, a spin density wave (SDW) state can be induced by Fermi surface nesting and electron-electron interaction. It may intertwine with other orders such as charge density wave (CDW), while their relation is still yet to be understood. Here via spin-polarized scanning tunneling microscopy, we directly observed long-range spin modulation on Cr(001) surface, which corresponds to the well-known incommensurate SDW of bulk Cr. It displays 6.0 nm in-plane period and anti-phase behavior between adjacent (001) planes. Meanwhile, we simultaneously observed the coexisting CDW with half the period of SDW. Such SDW/CDW have highly correlated domain structures and are in-phase. Surprisingly, the CDW displays a contrast inversion around a density-of-states dip at -22 meV, indicating an anomalous CDW gap opened below EF. These observations support that the CDW is a secondary order driven by SDW. Our work is not only a real-space characterization of incommensurate SDW, but also provides insights on how SDW and CDW coexist.

cond-mat.mtrl-sci↗

arXiv2022

Strong convergence rate of the Euler scheme for SDEs driven by additive rough fractional noises

The strong convergence rate of the Euler scheme for SDEs driven by additive fractional Brownian motions is studied, where the fractional Brownian motion has Hurst parameter $H\in(\frac13,\frac12)$ and the drift coefficient is not required to be bounded. The Malliavin calculus, the rough path theory and the $2$D Young integral are utilized to overcome the difficulties caused by the low regularity of the fractional Brownian motion and the unboundedness of the drift coefficient. The Euler scheme is proved to have strong order $2H$ for the case that the drift coefficient has bounded derivatives up to order three and have strong order $H+\frac12$ for linear cases. Numerical simulations are presented to support the theoretical results.

math.NA↗