arXiv subjects

Shuai Guo

Publications and source records attributed to Shuai Guo.

At least 37 records · Page 2Linked to original sources

arXiv2025

FF-PNet: A Pyramid Network Based on Feature and Field for Brain Image Registration

In recent years, deformable medical image registration techniques have made significant progress. However, existing models still lack efficiency in parallel extraction of coarse and fine-grained features. To address this, we construct a new pyramid registration network based on feature and deformation field (FF-PNet). For coarse-grained feature extraction, we design a Residual Feature Fusion Module (RFFM), for fine-grained image deformation, we propose a Residual Deformation Field Fusion Module (RDFFM). Through the parallel operation of these two modules, the model can effectively handle complex image deformations. It is worth emphasizing that the encoding stage of FF-PNet only employs traditional convolutional neural networks without any attention mechanisms or multilayer perceptrons, yet it still achieves remarkable improvements in registration accuracy, fully demonstrating the superior feature decoding capabilities of RFFM and RDFFM. We conducted extensive experiments on the LPBA and OASIS datasets. The results show our network consistently outperforms popular methods in metrics like the Dice Similarity Coefficient.

cs.CV↗

arXiv2025

Tetrahedron-Net for Medical Image Registration

Medical image registration plays a vital role in medical image processing. Extracting expressive representations for medical images is crucial for improving the registration quality. One common practice for this end is constructing a convolutional backbone to enable interactions with skip connections among feature extraction layers. The de facto structure, U-Net-like networks, has attempted to design skip connections such as nested or full-scale ones to connect one single encoder and one single decoder to improve its representation capacity. Despite being effective, it still does not fully explore interactions with a single encoder and decoder architectures. In this paper, we embrace this observation and introduce a simple yet effective alternative strategy to enhance the representations for registrations by appending one additional decoder. The new decoder is designed to interact with both the original encoder and decoder. In this way, it not only reuses feature presentation from corresponding layers in the encoder but also interacts with the original decoder to corporately give more accurate registration results. The new architecture is concise yet generalized, with only one encoder and two decoders forming a ``Tetrahedron'' structure, thereby dubbed Tetrahedron-Net. Three instantiations of Tetrahedron-Net are further constructed regarding the different structures of the appended decoder. Our extensive experiments prove that superior performance can be obtained on several representative benchmarks of medical image registration. Finally, such a ``Tetrahedron'' design can also be easily integrated into popular U-Net-like architectures including VoxelMorph, ViT-V-Net, and TransMorph, leading to consistent performance gains.

eess.IV↗

arXiv2025

Dynamic Backtracking in GFlowNets: Enhancing Decision Steps with Reward-Dependent Adjustment Mechanisms

Generative Flow Networks (GFlowNets or GFNs) are probabilistic models predicated on Markov flows, and they employ specific amortization algorithms to learn stochastic policies that generate compositional substances including biomolecules, chemical materials, etc. With a strong ability to generate high-performance biochemical molecules, GFNs accelerate the discovery of scientific substances, effectively overcoming the time-consuming, labor-intensive, and costly shortcomings of conventional material discovery methods. However, previous studies rarely focus on accumulating exploratory experience by adjusting generative structures, which leads to disorientation in complex sampling spaces. Efforts to address this issue, such as LS-GFN, are limited to local greedy searches and lack broader global adjustments. This paper introduces a novel variant of GFNs, the Dynamic Backtracking GFN (DB-GFN), which improves the adaptability of decision-making steps through a reward-based dynamic backtracking mechanism. DB-GFN allows backtracking during the network construction process according to the current state's reward value, thereby correcting disadvantageous decisions and exploring alternative pathways during the exploration process. When applied to generative tasks involving biochemical molecules and genetic material sequences, DB-GFN outperforms GFN models such as LS-GFN and GTB, as well as traditional reinforcement learning methods, in sample quality, sample exploration quantity, and training convergence speed. Additionally, owing to its orthogonal nature, DB-GFN shows great potential in future improvements of GFNs, and it can be integrated with other strategies to achieve higher search performance.

cs.LG↗

arXiv2025

Poincaré polynomials of moduli spaces of one-dimensional sheaves on the projective plane

Let $M_β$ denote the moduli space of stable one-dimensional sheaves on a del Pezzo surface $S$, supported on curves of class $β$ with Euler characteristic one. We show that the divisibility property of the Poincaré polynomial of $M_β$, proposed by Choi-van Garrel-Katz-Takahashi follows from Bousseau's conjectural refined sheaves/Gromov-Witten correspondence. Since this correspondence is known for $S=\mathbb{P}^2$, our result proves Choi-van Garrel-Katz-Takahashi's conjecture in this case. For $S=\mathbb{P}^2$, our proof also introduces a novel approach to computing the Poincaré polynomials using Gromov-Witten invariants of local $\mathbb{P}^2$ and a local elliptic curve. Specifically, we compute the Poincaré polynomials of $M_{d}$ with degrees $d\leq 16$ and derive a closed formula for the leading Betti numbers $b_i(M_d)$ with $d\geq 6$ and $i\leq 4d-22$. We also propose a conjectural formula for the leading Betti numbers $b_i(M_d)$ with $d\geq 4$ and $i\leq 6d-20$. In the Appendix (by M. Moreira), a more general conjecture concerning the higher range Betti numbers of $M_{d}$ is presented, along with another conjecture that involves refinements from the perverse/Chern filtration.

math.AG↗

arXiv2025

Cohomological Field Theory with vacuum and its Virasoro constraints

This is the first part of a series of papers on {\it Virasoro constraints for Cohomological Field Theory (CohFT)}. For a CohFT with vacuum, we introduce the concepts of $S$-calibration and $ν$-calibration. Then, we define the (formal) total descendent potential corresponding to a given calibration. Finally, we introduce an additional structure, namely homogeneity, for both the CohFT and the calibrations. After these preliminary introductions, we propose two crucial conjectures: (1) the ancestor version of the Virasoro conjecture for the homogeneous CohFT with vacuum; and (2) the generalized Virasoro conjecture for the (formal) total descendent potential of a calibrated homogeneous CohFT. We verify the genus-0 part of these conjectures and deduce a simplified form of the genus-1 part of these conjectures for arbitrary CohFTs. Additionally, we prove the full conjectures for semisimple CohFTs. As applications, our results yield the Virasoro constraints for the deformed negative $r$-spin theory. Moreover, by applying the Virasoro constraints, we discover an extension of Grothendieck's dessins d'enfants theory which is widely studied in the literature.

math-ph↗

arXiv2025

Sewer Image Super-Resolution with Depth Priors and Its Lightweight Network

The Quick-view (QV) technique serves as a primary method for detecting defects within sewerage systems. However, the effectiveness of QV is impeded by the limited visual range of its hardware, resulting in suboptimal image quality for distant portions of the sewer network. Image super-resolution is an effective way to improve image quality and has been applied in a variety of scenes. However, research on super-resolution for sewer images remains considerably unexplored. In response, this study leverages the inherent depth relationships present within QV images and introduces a novel Depth-guided, Reference-based Super-Resolution framework denoted as DSRNet. It comprises two core components: a depth extraction module and a depth information matching module (DMM). DSRNet utilizes the adjacent frames of the low-resolution image as reference images and helps them recover texture information based on the correlation. By combining these modules, the integration of depth priors significantly enhances both visual quality and performance benchmarks. Besides, in pursuit of computational efficiency and compactness, a super-resolution knowledge distillation model based on an attention mechanism is introduced. This mechanism facilitates the acquisition of feature similarity between a more complex teacher model and a streamlined student model, with the latter being a lightweight version of DSRNet. Experimental results demonstrate that DSRNet significantly improves PSNR and SSIM compared with other methods. This study also conducts experiments on sewer defect semantic segmentation, object detection, and classification on the Pipe dataset and Sewer-ML dataset. Experiments show that the method can improve the performance of low-resolution sewer images in these tasks.

cs.CV↗

arXiv2025

The bilinear fermionic form for KP and BKP hierarchies

For a tau-function of the KP or BKP hierarchy, we introduce the notion of lifting operator and derive an equation connecting the corresponding fermionic two-point function and fermionic one-point function through the lifting operator. This provides an effective approach to determine the fermionic two-point function of the tau-function from the lifting operator and the fermionic one-point function. As practical applications, we derive concise formulas for the fermionic two-point functions of several models, like the $r$-spin model and the Br{\' e}zin--Gross--Witten model, which respectively serve as examples for KP and BKP tau-functions.

math-ph↗

arXiv2025

Online Robot Motion Planning Methodology Guided by Group Social Proxemics Feature

Nowadays robot is supposed to demonstrate human-like perception, reasoning and behavior pattern in social or service application. However, most of the existing motion planning methods are incompatible with above requirement. A potential reason is that the existing navigation algorithms usually intend to treat people as another kind of obstacle, and hardly take the social principle or awareness into consideration. In this paper, we attempt to model the proxemics of group and blend it into the scenario perception and navigation of robot. For this purpose, a group clustering method considering both social relevance and spatial confidence is introduced. It can enable robot to identify individuals and divide them into groups. Next, we propose defining the individual proxemics within magnetic dipole model, and further established the group proxemics and scenario map through vector-field superposition. On the basis of the group clustering and proxemics modeling, we present the method to obtain the optimal observation positions (OOPs) of group. Once the OOPs grid and scenario map are established, a heuristic path is employed to generate path that guide robot cruising among the groups for interactive purpose. A series of experiments are conducted to validate the proposed methodology on the practical robot, the results have demonstrated that our methodology has achieved promising performance on group recognition accuracy and path-generation efficiency. This concludes that the group awareness evolved as an important module to make robot socially behave in the practical scenario.

cs.RO↗

arXiv2024

A New People-Object Interaction Dataset and NVS Benchmarks

Recently, NVS in human-object interaction scenes has received increasing attention. Existing human-object interaction datasets mainly consist of static data with limited views, offering only RGB images or videos, mostly containing interactions between a single person and objects. Moreover, these datasets exhibit complexities in lighting environments, poor synchronization, and low resolution, hindering high-quality human-object interaction studies. In this paper, we introduce a new people-object interaction dataset that comprises 38 series of 30-view multi-person or single-person RGB-D video sequences, accompanied by camera parameters, foreground masks, SMPL models, some point clouds, and mesh files. Video sequences are captured by 30 Kinect Azures, uniformly surrounding the scene, each in 4K resolution 25 FPS, and lasting for 1$\sim$19 seconds. Meanwhile, we evaluate some SOTA NVS models on our dataset to establish the NVS benchmarks. We hope our work can inspire further research in humanobject interaction.

cs.CV↗

arXiv2024

Depth-Guided Robust and Fast Point Cloud Fusion NeRF for Sparse Input Views

Novel-view synthesis with sparse input views is important for real-world applications like AR/VR and autonomous driving. Recent methods have integrated depth information into NeRFs for sparse input synthesis, leveraging depth prior for geometric and spatial understanding. However, most existing works tend to overlook inaccuracies within depth maps and have low time efficiency. To address these issues, we propose a depth-guided robust and fast point cloud fusion NeRF for sparse inputs. We perceive radiance fields as an explicit voxel grid of features. A point cloud is constructed for each input view, characterized within the voxel grid using matrices and vectors. We accumulate the point cloud of each input view to construct the fused point cloud of the entire scene. Each voxel determines its density and appearance by referring to the point cloud of the entire scene. Through point cloud fusion and voxel grid fine-tuning, inaccuracies in depth values are refined or substituted by those from other views. Moreover, our method can achieve faster reconstruction and greater compactness through effective vector-matrix decomposition. Experimental results underline the superior performance and time efficiency of our approach compared to state-of-the-art baselines.

cs.CV↗

arXiv2024

Joint Attention-Guided Feature Fusion Network for Saliency Detection of Surface Defects

Surface defect inspection plays an important role in the process of industrial manufacture and production. Though Convolutional Neural Network (CNN) based defect inspection methods have made huge leaps, they still confront a lot of challenges such as defect scale variation, complex background, low contrast, and so on. To address these issues, we propose a joint attention-guided feature fusion network (JAFFNet) for saliency detection of surface defects based on the encoder-decoder network. JAFFNet mainly incorporates a joint attention-guided feature fusion (JAFF) module into decoding stages to adaptively fuse low-level and high-level features. The JAFF module learns to emphasize defect features and suppress background noise during feature fusion, which is beneficial for detecting low-contrast defects. In addition, JAFFNet introduces a dense receptive field (DRF) module following the encoder to capture features with rich context information, which helps detect defects of different scales. The JAFF module mainly utilizes a learned joint channel-spatial attention map provided by high-level semantic features to guide feature fusion. The attention map makes the model pay more attention to defect features. The DRF module utilizes a sequence of multi-receptive-field (MRF) units with each taking as inputs all the preceding MRF feature maps and the original input. The obtained DRF features capture rich context information with a large range of receptive fields. Extensive experiments conducted on SD-saliency-900, Magnetic tile, and DAGM 2007 indicate that our method achieves promising performance in comparison with other state-of-the-art methods. Meanwhile, our method reaches a real-time defect detection speed of 66 FPS.

cs.CV↗

arXiv2024

Self-supervised Learning for Electroencephalogram: A Systematic Survey

Electroencephalogram (EEG) is a non-invasive technique to record bioelectrical signals. Integrating supervised deep learning techniques with EEG signals has recently facilitated automatic analysis across diverse EEG-based tasks. However, the label issues of EEG signals have constrained the development of EEG-based deep models. Obtaining EEG annotations is difficult that requires domain experts to guide collection and labeling, and the variability of EEG signals among different subjects causes significant label shifts. To solve the above challenges, self-supervised learning (SSL) has been proposed to extract representations from unlabeled samples through well-designed pretext tasks. This paper concentrates on integrating SSL frameworks with temporal EEG signals to achieve efficient representation and proposes a systematic review of the SSL for EEG signals. In this paper, 1) we introduce the concept and theory of self-supervised learning and typical SSL frameworks. 2) We provide a comprehensive review of SSL for EEG analysis, including taxonomy, methodology, and technique details of the existing EEG-based SSL frameworks, and discuss the difference between these methods. 3) We investigate the adaptation of the SSL approach to various downstream tasks, including the task description and related benchmark datasets. 4) Finally, we discuss the potential directions for future SSL-EEG research.

eess.SP↗

arXiv2023

Average Jaccard Index of Random Graphs

The asymptotic behavior of the Jaccard index in $G(n,p)$, the classical Erdös-Rényi random graphs model, is studied in this paper, as $n$ goes to infinity. We first derive the asymptotic distribution of the Jaccard index of any pair of distinct vertices, as well as the first two moments of this index. Then the average of the Jaccard indices over all vertex pairs in $G(n,p)$ is shown to be asymptotically normal under an additional mild condition that $np\to\infty$ and $n^2(1-p)\to\infty$.

math.PR↗

arXiv2023

Multi-GradSpeech: Towards Diffusion-based Multi-Speaker Text-to-speech Using Consistent Diffusion Models

Despite imperfect score-matching causing drift in training and sampling distributions of diffusion models, recent advances in diffusion-based acoustic models have revolutionized data-sufficient single-speaker Text-to-Speech (TTS) approaches, with Grad-TTS being a prime example. However, the sampling drift problem leads to these approaches struggling in multi-speaker scenarios in practice due to more complex target data distribution compared to single-speaker scenarios. In this paper, we present Multi-GradSpeech, a multi-speaker diffusion-based acoustic models which introduces the Consistent Diffusion Model (CDM) as a generative modeling approach. We enforce the consistency property of CDM during the training process to alleviate the sampling drift problem in the inference stage, resulting in significant improvements in multi-speaker TTS performance. Our experimental results corroborate that our proposed approach can improve the performance of different speakers involved in multi-speaker TTS compared to Grad-TTS, even outperforming the fine-tuning approach. Audio samples are available at https://welkinyang.github.io/multi-gradspeech/

eess.AS↗

arXiv2022

Exponential canonical correlation analysis with orthogonal variation

Canonical correlation analysis (CCA) is a standard tool for studying associations between two data sources; however, it is not designed for data with count or proportion measurement types. In addition, while CCA uncovers common signals, it does not elucidate which signals are unique to each data source. To address these challenges, we propose a new framework for CCA based on exponential families with explicit modeling of both common and source-specific signals. Unlike previous methods based on exponential families, the common signals from our model coincide with canonical variables in Gaussian CCA, and the unique signals are exactly orthogonal. These modeling differences lead to a non-trivial estimation via optimization with orthogonality constraints, for which we develop an iterative algorithm based on a splitting method. Simulations show on par or superior performance of the proposed method compared to the available alternatives. We apply the method to analyze associations between gene expressions and lipids concentrations in nutrigenomic study, and to analyze associations between two distinct cell-type deconvolution methods in prostate cancer tumor heterogeneity study.

stat.CO↗

arXiv2022

SingAug: Data Augmentation for Singing Voice Synthesis with Cycle-consistent Training Strategy

Deep learning based singing voice synthesis (SVS) systems have been demonstrated to flexibly generate singing with better qualities, compared to conventional statistical parametric based methods. However, neural systems are generally data-hungry and have difficulty to reach reasonable singing quality with limited public available training data. In this work, we explore different data augmentation methods to boost the training of SVS systems, including several strategies customized to SVS based on pitch augmentation and mix-up augmentation. To further stabilize the training, we introduce the cycle-consistent training strategy. Extensive experiments on two public singing databases demonstrate that our proposed augmentation methods and the stabilizing training strategy can significantly improve the performance on both objective and subjective evaluations.

eess.AS↗

arXiv2022

Muskits: an End-to-End Music Processing Toolkit for Singing Voice Synthesis

This paper introduces a new open-source platform named Muskits for end-to-end music processing, which mainly focuses on end-to-end singing voice synthesis (E2E-SVS). Muskits supports state-of-the-art SVS models, including RNN SVS, transformer SVS, and XiaoiceSing. The design of Muskits follows the style of widely-used speech processing toolkits, ESPnet and Kaldi, for data prepossessing, training, and recipe pipelines. To the best of our knowledge, this toolkit is the first platform that allows a fair and highly-reproducible comparison between several published works in SVS. In addition, we also demonstrate several advanced usages based on the toolkit functionalities, including multilingual training and transfer learning. This paper describes the major framework of Muskits, its functionalities, and experimental results in single-singer, multi-singer, multilingual, and transfer learning scenarios. The toolkit is publicly available at https://github.com/SJTMusicTeam/Muskits.

cs.SD↗

arXiv2022

The low-entropy hydration shell at the binding site of spike RBD determines the contagiousness of SARS-CoV-2 variants

The infectivity of SARS-CoV-2 depends on the binding affinity of the receptor-binding domain (RBD) of the spike protein with the angiotensin converting enzyme 2 (ACE2) receptor. The calculated RBD-ACE2 binding energies indicate that the difference in transmission efficiency of SARS-CoV-2 variants cannot be fully explained by electrostatic interactions, hydrogen-bond interactions, van der Waals interactions, internal energy, and nonpolar solvation energies. Here, we demonstrate that low-entropy regions of hydration shells around proteins drive hydrophobic attraction between shape-matched low-entropy regions of the hydration shells, which essentially coordinates protein-protein binding in rotational-configurational space of mutual orientations and determines the binding affinity. An innovative method was used to identify the low-entropy regions of the hydration shells of the RBDs of multiple SARS-CoV-2 variants and the ACE2. We observed integral low-entropy regions of hydration shells covering the binding sites of the RBDs and matching in shape to the low-entropy region of hydration shell at the binding site of the ACE2. The RBD-ACE2 binding is thus found to be guided by hydrophobic collapse between the shape-matched low-entropy regions of the hydration shells. A measure of the low-entropy of the hydration shells can be obtained by counting the number of hydrophilic groups expressing hydrophilicity within the binding sites. The low-entropy level of hydration shells at the binding site of a spike protein is found to be an important indicator of the contagiousness of the coronavirus.

q-bio.BM↗