Search

Modern semantic segmentation methods devote much effect to adjusting image feature representations to improve the segmentation performance in various ways, such as architecture design, attention mechnism, etc. However, almost all those methods neglect the particularity of class weights (in the classification layer) in segmentation models. In this paper, we notice that the class weights of categories that tend to share many adjacent boundary pixels lack discrimination, thereby limiting the performance. We call this issue Boundary-caused Class Weights Confusion (BCWC). We try to focus on this problem and propose a novel method named Embedded Conditional Random Field (E-CRF) to alleviate it. E-CRF innovatively fuses the CRF into the CNN network as an organic whole for more effective end-to-end optimization. The reasons are two folds. It utilizes CRF to guide the message passing between pixels in high-level features to purify the feature representation of boundary pixels, with the help of inner pixels belonging to the same object. More importantly, it enables optimizing class weights from both scale and direction during backpropagation. We make detailed theoretical analysis to prove it. Besides, superpixel is integrated into E-CRF and served as an auxiliary to exploit the local object prior for more reliable message passing. Finally, our proposed method yields impressive results on ADE20K, Cityscapes, and Pascal Context datasets., Comment: Accepted by ICLR2023. Camera-ready Version

In this paper, we propose a simple but effective message passing method to improve the boundary quality for the semantic segmentation result. Inspired by the generated sharp edges of superpixel blocks, we employ superpixel to guide the information passing within feature map. Simultaneously, the sharp boundaries of the blocks also restrict the message passing scope. Specifically, we average features that the superpixel block covers within feature map, and add the result back to each feature vector. Further, to obtain sharper edges and farther spatial dependence, we develop a multiscale superpixel module (MSP) by a cascade of different scales superpixel blocks. Our method can be served as a plug-and-play module and easily inserted into any segmentation network without introducing new parameters. Extensive experiments are conducted on three strong baselines, namely PSPNet, DeeplabV3, and DeepLabV3+, and four challenging scene parsing datasets including ADE20K, Cityscapes, PASCAL VOC, and PASCAL Context. The experimental results verify its effectiveness and generalizability., Comment: under review

This paper proposes a network, referred to as MVSTR, for Multi-View Stereo (MVS). It is built upon Transformer and is capable of extracting dense features with global context and 3D consistency, which are crucial to achieving reliable matching for MVS. Specifically, to tackle the problem of the limited receptive field of existing CNN-based MVS methods, a global-context Transformer module is first proposed to explore intra-view global context. In addition, to further enable dense features to be 3D-consistent, a 3D-geometry Transformer module is built with a well-designed cross-view attention mechanism to facilitate inter-view information interaction. Experimental results show that the proposed MVSTR achieves the best overall performance on the DTU dataset and strong generalization on the Tanks & Temples benchmark dataset.

Topology-driven nonlinear light-matter effects open up new paradigms for both topological photonics and nonlinear optics. Here, we propose to achieve high-efficiency second-harmonic generation in a second-order photonic topological insulator. Such system hosts highly localized topological corner states with large quality factors for both fundamental and second harmonic waves, which could be matched perfectly in frequency by simply tuning the structural parameters. Through the nonlinear interaction of the doubly resonant topological corner states, unprecedented frequency conversion efficiency is predicted. In addition, the robustness of the nonlinear process against defects is also demonstrated. Our work opens up new avenues toward topologically protected nonlinear frequency conversion using high-order photonic topological modes., Comment: 7 pages, 8 figures

Artificial intelligence (AI) technology is increasingly used for digital orthodontics, but one of the challenges is to automatically and accurately detect tooth landmarks and axes. This is partly because of sophisticated geometric definitions of them, and partly due to large variations among individual tooth and across different types of tooth. As such, we propose a deep learning approach with a labeled dataset by professional dentists to the tooth landmark/axis detection on tooth model that are crucial for orthodontic treatments. Our method can extract not only tooth landmarks in the form of point (e.g. cusps), but also axes that measure the tooth angulation and inclination. The proposed network takes as input a 3D tooth model and predicts various types of the tooth landmarks and axes. Specifically, we encode the landmarks and axes as dense fields defined on the surface of the tooth model. This design choice and a set of added components make the proposed network more suitable for extracting sparse landmarks from a given 3D tooth model. Extensive evaluation of the proposed method was conducted on a set of dental models prepared by experienced dentists. Results show that our method can produce tooth landmarks with high accuracy. Our method was examined and justified via comparison with the state-of-the-art methods as well as the ablation studies., Comment: 11pages,27figures

Thouless pump is a one-dimensional dynamic topological effect that stems from the same topological mechanism as the renowned two-dimensional Chern insulators, with one momentum dimension replaced by a time variant evolution parameter. The underlying physics is abelian, concerning topologies of individual bands. By introducing multi-band non-abelian topology into pumping, much richer physics is expected that goes beyond the abelian counterpart. Here we report the observation of a non-abelian Thouless pump in an acoustic waveguide array with judiciously engineered coupling configurations. The observed non-abelian effect originates from time-evolution of eigen-states in Z_3 pump cycles that traverse across several band degeneracies in the parameter space in a three-band system. Interestingly, the pump of the eigenstates exhibits non-commutativity, i.e. the final state depends critically on the order of the pumping paths. Our study paves ways for exploring and utilizing non-abelian dynamical effects in classical systems., Comment: 9 pages, 5 figures

Tibetan is a low-resource language. In order to alleviate the shortage of parallel corpus between Tibetan and Chinese, this paper uses two monolingual corpora and a small number of seed dictionaries to learn the semi-supervised method with seed dictionaries and self-supervised adversarial training method through the similarity calculation of word clusters in different embedded spaces and puts forward an improved self-supervised adversarial learning method of Tibetan and Chinese monolingual data alignment only. The experimental results are as follows. First, the experimental results of Tibetan syllables Chinese characters are not good, which reflects the weak semantic correlation between Tibetan syllables and Chinese characters; second, the seed dictionary of semi-supervised method made before 10 predicted word accuracy of 66.5 (Tibetan - Chinese) and 74.8 (Chinese - Tibetan) results, to improve the self-supervision methods in both language directions have reached 53.5 accuracy., Comment: 8pages,2nd International Conference on Machine Learning Techniques and NLP (MLNLP 2021)

Recently, there is great interest to investigate the application of deep learning models for the prediction of clinical events using electronic health records (EHR) data. In EHR data, a patient's history is often represented as a sequence of visits, and each visit contains multiple events. As a result, deep learning models developed for sequence modeling, like recurrent neural networks (RNNs) are common architecture for EHR-based clinical events predictive models. While a large variety of RNN models were proposed in the literature, it is unclear if complex architecture innovations will offer superior predictive performance. In order to move this field forward, a rigorous evaluation of various methods is needed. In this study, we conducted a thorough benchmark of RNN architectures in modeling EHR data. We used two prediction tasks: the risk for developing heart failure and the risk of early readmission for inpatient hospitalization. We found that simple gated RNN models, including GRUs and LSTMs, often offer competitive results when properly tuned with Bayesian Optimization, which is in line with similar to findings in the natural language processing (NLP) domain. For reproducibility, Our codebase is shared at https://github.com/ZhiGroup/pytorch_ehr., Comment: This podium abstract was peer-reviewed and accepted in 2018, and presented at Medinfo-2019 in Leon France in August 2019

We propose a general formulation for continuous treatment recommendation problems in settings with clinical survival data, which we call the Deep Survival Dose Response Function (DeepSDRF). That is, we consider the problem of learning the conditional average dose response (CADR) function solely from historical data in which observed factors (confounders) affect both observed treatment and time-to-event outcomes. The estimated treatment effect from DeepSDRF enables us to develop recommender algorithms with the correction for selection bias. We compared two recommender approaches based on random search and reinforcement learning and found similar performance in terms of patient outcome. We tested the DeepSDRF and the corresponding recommender on extensive simulation studies and the eICU Research Institute (eRI) database. To the best of our knowledge, this is the first time that causal models are used to address the continuous treatment effect with observational data in a medical context., Comment: results is outdated

Previous researches on high-energy neutrino events from gamma-ray bursters (GRBs) suggest a neutrino speed variation $v(E)=c(1\pm E/E^{\nu}_{\mathrm{LV}})$ with ${E}^{\nu}_{\rm LV}=(6.4\pm 1.5)\times10^{17}~{ \rm GeV}$, together with an intrinsic time difference ${\Delta {t}_{\rm in}=(-2.8\pm 0.7)\times10^2~{\rm s}}$, which means that high-energy neutrinos come out about 300~s earlier than low-energy photons in the source reference system. Considering the possibility that pre-bursts of neutrinos may be accompanied by high-energy photons, in this work we search for high-energy photon events with earlier emission time from 100 to 1000~s before low-energy photons at source by analyzing Fermi Gamma-ray Space Telescope (FGST) data. We perform the searching of photon events with energies larger than 100~MeV, and find 14 events from 48 GRBs with known redshifts. Combining these events with a $1.07~\rm{TeV}$ photon event observed by the Major Atmospheric Gamma Imaging Cherenkov telescopes (MAGIC), we suggest a pre-burst stage with a long duration period of several minutes of high energy neutrino emissions accompanied by high energy photons at the GRB source., Comment: 7 latex pages, 1 figure, final version for publication