arXiv subjects

Ming Cai

Publications and source records attributed to Ming Cai.

29 records · Page 2Linked to original sources

arXiv2022

FCGEC: Fine-Grained Corpus for Chinese Grammatical Error Correction

Grammatical Error Correction (GEC) has been broadly applied in automatic correction and proofreading system recently. However, it is still immature in Chinese GEC due to limited high-quality data from native speakers in terms of category and scale. In this paper, we present FCGEC, a fine-grained corpus to detect, identify and correct the grammatical errors. FCGEC is a human-annotated corpus with multiple references, consisting of 41,340 sentences collected mainly from multi-choice questions in public school Chinese examinations. Furthermore, we propose a Switch-Tagger-Generator (STG) baseline model to correct the grammatical errors in low-resource settings. Compared to other GEC benchmark models, experimental results illustrate that STG outperforms them on our FCGEC. However, there exists a significant gap between benchmark models and humans that encourages future models to bridge it.

cs.CL↗

arXiv2022

Learn from Structural Scope: Improving Aspect-Level Sentiment Analysis with Hybrid Graph Convolutional Networks

Aspect-level sentiment analysis aims to determine the sentiment polarity towards a specific target in a sentence. The main challenge of this task is to effectively model the relation between targets and sentiments so as to filter out noisy opinion words from irrelevant targets. Most recent efforts capture relations through target-sentiment pairs or opinion spans from a word-level or phrase-level perspective. Based on the observation that targets and sentiments essentially establish relations following the grammatical hierarchy of phrase-clause-sentence structure, it is hopeful to exploit comprehensive syntactic information for better guiding the learning process. Therefore, we introduce the concept of Scope, which outlines a structural text region related to a specific target. To jointly learn structural Scope and predict the sentiment polarity, we propose a hybrid graph convolutional network (HGCN) to synthesize information from constituency tree and dependency tree, exploring the potential of linking two syntax parsing methods to enrich the representation. Experimental results on four public datasets illustrate that our HGCN model outperforms current state-of-the-art baselines.

cs.CL↗

arXiv2021

Graph-based Pyramid Global Context Reasoning with a Saliency-aware Projection for COVID-19 Lung Infections Segmentation

Coronavirus Disease 2019 (COVID-19) has rapidly spread in 2020, emerging a mass of studies for lung infection segmentation from CT images. Though many methods have been proposed for this issue, it is a challenging task because of infections of various size appearing in different lobe zones. To tackle these issues, we propose a Graph-based Pyramid Global Context Reasoning (Graph-PGCR) module, which is capable of modeling long-range dependencies among disjoint infections as well as adapt size variation. We first incorporate graph convolution to exploit long-term contextual information from multiple lobe zones. Different from previous average pooling or maximum object probability, we propose a saliency-aware projection mechanism to pick up infection-related pixels as a set of graph nodes. After graph reasoning, the relation-aware features are reversed back to the original coordinate space for the down-stream tasks. We further construct multiple graphs with different sampling rates to handle the size variation problem. To this end, distinct multi-scale long-range contextual patterns can be captured. Our Graph-PGCR module is plug-and-play, which can be integrated into any architecture to improve its performance. Experiments demonstrated that the proposed method consistently boost the performance of state-of-the-art backbone architectures on both of public and our private COVID-19 datasets.

eess.IV↗

arXiv2021

PA-ResSeg: A Phase Attention Residual Network for Liver Tumor Segmentation from Multi-phase CT Images

In this paper, we propose a phase attention residual network (PA-ResSeg) to model multi-phase features for accurate liver tumor segmentation, in which a phase attention (PA) is newly proposed to additionally exploit the images of arterial (ART) phase to facilitate the segmentation of portal venous (PV) phase. The PA block consists of an intra-phase attention (Intra-PA) module and an inter-phase attention (Inter-PA) module to capture channel-wise self-dependencies and cross-phase interdependencies, respectively. Thus it enables the network to learn more representative multi-phase features by refining the PV features according to the channel dependencies and recalibrating the ART features based on the learned interdependencies between phases. We propose a PA-based multi-scale fusion (MSF) architecture to embed the PA blocks in the network at multiple levels along the encoding path to fuse multi-scale features from multi-phase images. Moreover, a 3D boundary-enhanced loss (BE-loss) is proposed for training to make the network more sensitive to boundaries. To evaluate the performance of our proposed PA-ResSeg, we conducted experiments on a multi-phase CT dataset of focal liver lesions (MPCT-FLLs). Experimental results show the effectiveness of the proposed method by achieving a dice per case (DPC) of 0.77.87, a dice global (DG) of 0.8682, a volumetric overlap error (VOE) of 0.3328 and a relative volume difference (RVD) of 0.0443 on the MPCT-FLLs. Furthermore, to validate the effectiveness and robustness of PA-ResSeg, we conducted extra experiments on another multi-phase liver tumor dataset and obtained a DPC of 0.8290, a DG of 0.9132, a VOE of 0.2637 and a RVD of 0.0163. The proposed method shows its robustness and generalization capability in different datasets and different backbones.

eess.IV↗

arXiv2020

P-Cloth: Interactive Complex Cloth Simulation on Multi-GPU Systems using Dynamic Matrix Assembly and Pipelined Implicit Integrators

We present a novel parallel algorithm for cloth simulation that exploits multiple GPUs for fast computation and the handling of very high resolution meshes. To accelerate implicit integration, we describe new parallel algorithms for sparse matrix-vector multiplication (SpMV) and for dynamic matrix assembly on a multi-GPU workstation. Our algorithms use a novel work queue generation scheme for a fat-tree GPU interconnect topology. Furthermore, we present a novel collision handling scheme that uses spatial hashing for discrete and continuous collision detection along with a non-linear impact zone solver. Our parallel schemes can distribute the computation and storage overhead among multiple GPUs and enable us to perform almost interactive simulation on complex cloth meshes, which can hardly be handled on a single GPU due to memory limitations. We have evaluated the performance with two multi-GPU workstations (with 4 and 8 GPUs, respectively) on cloth meshes with 0.5-1.65M triangles. Our approach can reliably handle the collisions and generate vivid wrinkles and folds at 2-5 fps, which is significantly faster than prior cloth simulation systems. We observe almost linear speedups with respect to the number of GPUs.

cs.GR↗

arXiv2019

Single-view Object Shape Reconstruction Using Deep Shape Prior and Silhouette

3D shape reconstruction from a single image is a highly ill-posed problem. Modern deep learning based systems try to solve this problem by learning an end-to-end mapping from image to shape via a deep network. In this paper, we aim to solve this problem via an online optimization framework inspired by traditional methods. Our framework employs a deep autoencoder to learn a set of latent codes of 3D object shapes, which are fitted by a probabilistic shape prior using Gaussian Mixture Model (GMM). At inference, the shape and pose are jointly optimized guided by both image cues and deep shape prior without relying on an initialization from any trained deep nets. Surprisingly, our method achieves comparable performance to state-of-the-art methods even without training an end-to-end network, which shows a promising step in this direction.

cs.CV↗

arXiv2019

Tropical and Extratropical General Circulation with a Meridional Reversed Temperature Gradient as Expected in a High Obliquity Planet

Planets with high obliquity receive more radiation in the polar regions than at low latitudes, and thus, assuming an ocean-covered surface with sufficiently high heat capacity, their meridional temperature gradient was shown to be reversed for the entire year. The objective of this work is to investigate the drastically different general circulation of such planets, with an emphasis on the tropical Hadley circulation and the mid-latitude baroclinic eddy structure. We use a 3D dry dynamic core model, accompanied by an eddy-free configuration and a generalized 2D Eady model. When the meridional temperature gradient $T_y$ is reversed, the Hadley cell remains in the same direction, because the surface wind pattern and hence the associated meridional Ekman transport are not changed, as required by the baroclinic eddy momentum transport. The Hadley cell under reversed $T_y$ also becomes much shallower and weaker, even when the magnitude of the gradient is the same as in the normal case. The shallowness is due to the bottom-heavy structure of the baroclinic eddies in the reverse case, and the weakness is due to the weak wave activity. We propose a new mechanism to explain the mid-latitude eddy structure for both cases, and verify it using the generalized Eady model. With seasonal variations included, the annual mean circulation resembles that under perpetual annual mean setup. Approaching the solstices, a strong cross-equator Hadley cell forms in both cases, and about 2/3 of the Hadley circulation is driven by eddies, as shown by eddy-free simulations and using a decomposition of the Hadley cell.

astro-ph.EP↗

arXiv2018

Attentive Sequence to Sequence Translation for Localizing Clips of Interest by Natural Language Descriptions

We propose a novel attentive sequence to sequence translator (ASST) for clip localization in videos by natural language descriptions. We make two contributions. First, we propose a bi-directional Recurrent Neural Network (RNN) with a finely calibrated vision-language attentive mechanism to comprehensively understand the free-formed natural language descriptions. The RNN parses natural language descriptions in two directions, and the attentive model attends every meaningful word or phrase to each frame, thereby resulting in a more detailed understanding of video content and description semantics. Second, we design a hierarchical architecture for the network to jointly model language descriptions and video content. Given a video-description pair, the network generates a matrix representation, i.e., a sequence of vectors. Each vector in the matrix represents a video frame conditioned by the description. The 2D representation not only preserves the temporal dependencies of frames but also provides an effective way to perform frame-level video-language matching. The hierarchical architecture exploits video content with multiple granularities, ranging from subtle details to global context. Integration of the multiple granularities yields a robust representation for multi-level video-language abstraction. We validate the effectiveness of our ASST on two large-scale datasets. Our ASST outperforms the state-of-the-art by $4.28\%$ in Rank$@1$ on the DiDeMo dataset. On the Charades-STA dataset, we significantly improve the state-of-the-art by $13.41\%$ in Rank$@1,IoU=0.5$.

cs.CV↗

arXiv2018

Deep-6DPose: Recovering 6D Object Pose from a Single RGB Image

Detecting objects and their 6D poses from only RGB images is an important task for many robotic applications. While deep learning methods have made significant progress in visual object detection and segmentation, the object pose estimation task is still challenging. In this paper, we introduce an end-toend deep learning framework, named Deep-6DPose, that jointly detects, segments, and most importantly recovers 6D poses of object instances from a single RGB image. In particular, we extend the recent state-of-the-art instance segmentation network Mask R-CNN with a novel pose estimation branch to directly regress 6D object poses without any post-refinements. Our key technical contribution is the decoupling of pose parameters into translation and rotation so that the rotation can be regressed via a Lie algebra representation. The resulting pose regression loss is differential and unconstrained, making the training tractable. The experiments on two standard pose benchmarking datasets show that our proposed approach compares favorably with the state-of-the-art RGB-based multi-stage pose estimation methods. Importantly, due to the end-to-end architecture, Deep-6DPose is considerably faster than competing multi-stage methods, offers an inference speed of 10 fps that is well suited for robotic applications.

cs.CV↗

arXiv2017

Baroclinic Instability and Transitions in a Two-Layer Quasigeostrophic Channel Model

The main objective of this article is to derive a mathematical theory associated with the nonlinear stability and dynamic transitions of the basic shear flows associated with baroclinic instability, which plays a fundamental role in the dominant mechanism shaping the cyclones and anticyclones that dominate weather in mid-latitudes, as well as the mesoscale eddies that play various roles in oceanic dynamics and the transport of tracers. This article provides a general transition and stability theory for the two-layer quasi-geostrophic model originally derived by Pedlosky \cite{ped1970}. We show that the instability and dynamic transition of the basic shear flow occur only when the Froude number $F>(γ^2+1)/2$, where $γ$ is the weave number of lowest zonal harmonic. In this case, we derive a precise critical shear velocity $U_c$ and a dynamic transition number $b$, and we show that the dynamic transition and associated flo0w patterns are dictated by the sign of this transition number. In particular, we show that for $b>0$, the system undergoes a continuous transition, leading to spatiotemporal flow oscillatory patterns as the shear velocity $U$ crosses $U_c$; for $b<0$, the system undergoes a jump transition, leading to drastic change and more complex transition patterns in the far field, and to unstable period solutions for $U<U_c$. Furthermore, we show that when the wavenumber of the zonal harmonic $γ=1$, the transition number $b$ is always positive and the system always undergoes a continuous dynamic transition leading spatiotemporal oscillations. This suggests that a continuous transition to spatiotemporal patterns is preferable for the shear flow associated with baroclinic instability.

physics.ao-ph↗

arXiv2015

Intermediate Coherent-entangled State Representation: Generation and its applications

By combining the beam splitter and the Fresnel transform, a protocol is proposed to generate a new entangled state representation, called the intermediate coherent-entangled state (ICES) representation. The properties, such as eigenvalue equation, completeness relation and orthogonal relation, are investigated. The conjugate state representation of the ICES and the Schmidt decomposing of the ICES are also discussed. As applications, a new squeezing operator and some operator identities by using the ICES are obtained.

quant-ph↗