Search

We study the sparse multi-type Erd\H{o}s R\'enyi random graphs. Despite that the corresponding central limit results are unknown, we are able to prove the moderate deviation principles for the size of the largest connected component, the number of a certain type of connected components, as well as the total number of the connected components. The rate functions are given explicitly. The proof relies on a careful study of a conditional multi-dimensional compound Poisson process. The properties of the related multi-type branching processes are also discussed.

This paper presents PlanarSplatting, an ultra-fast and accurate surface reconstruction approach for multiview indoor images. We take the 3D planes as the main objective due to their compactness and structural expressiveness in indoor scenes, and develop an explicit optimization framework that learns to fit the expected surface of indoor scenes by splatting the 3D planes into 2.5D depth and normal maps. As our PlanarSplatting operates directly on the 3D plane primitives, it eliminates the dependencies on 2D/3D plane detection and plane matching and tracking for planar surface reconstruction. Furthermore, the essential merits of plane-based representation plus CUDA-based implementation of planar splatting functions, PlanarSplatting reconstructs an indoor scene in 3 minutes while having significantly better geometric accuracy. Thanks to our ultra-fast reconstruction speed, the largest quantitative evaluation on the ScanNet and ScanNet++ datasets over hundreds of scenes clearly demonstrated the advantages of our method. We believe that our accurate and ultrafast planar surface reconstruction method will be applied in the structured data curation for surface reconstruction in the future. The code of our CUDA implementation will be publicly available. Project page: https://icetttb.github.io/PlanarSplatting/, Comment: Project page: https://icetttb.github.io/PlanarSplatting/

Assistive technologies for people with visual impairments (PVI) have made significant advancements, particularly with the integration of artificial intelligence (AI) and real-time sensor technologies. However, current solutions often require PVI to switch between multiple apps and tools for tasks like image recognition, navigation, and obstacle detection, which can hinder a seamless and efficient user experience. In this paper, we present NaviGPT, a high-fidelity prototype that integrates LiDAR-based obstacle detection, vibration feedback, and large language model (LLM) responses to provide a comprehensive and real-time navigation aid for PVI. Unlike existing applications such as Be My AI and Seeing AI, NaviGPT combines image recognition and contextual navigation guidance into a single system, offering continuous feedback on the user's surroundings without the need for app-switching. Meanwhile, NaviGPT compensates for the response delays of LLM by using location and sensor data, aiming to provide practical and efficient navigation support for PVI in dynamic environments., Comment: 7 pages, 3 figures, this work has been accepted by the 2025 ACM International Conference on Supporting Group Work (GROUP '25)

The LiDAR-based 3D object detector that strikes a balance between accuracy and speed is crucial for achieving real-time perception in autonomous driving and robotic navigation systems. To enhance the accuracy of point cloud detection, integrating global context for visual understanding improves the point clouds ability to grasp overall spatial information. However, many existing LiDAR detection models depend on intricate feature transformation and extraction processes, leading to poor real-time performance and high resource consumption, which limits their practical effectiveness. In this work, we propose a Faster LiDAR 3D object detection framework, called FASD, which implements heterogeneous model distillation by adaptively uniform cross-model voxel features. We aim to distill the transformer's capacity for high-performance sequence modeling into Mamba models with low FLOPs, achieving a significant improvement in accuracy through knowledge transfer. Specifically, Dynamic Voxel Group and Adaptive Attention strategies are integrated into the sparse backbone, creating a robust teacher model with scale-adaptive attention for effective global visual context modeling. Following feature alignment with the Adapter, we transfer knowledge from the Transformer to the Mamba through latent space feature supervision and span-head distillation, resulting in improved performance and an efficient student model. We evaluated the framework on the Waymo and nuScenes datasets, achieving a 4x reduction in resource consumption and a 1-2\% performance improvement over the current SoTA methods.

Accurate and robust LiDAR 3D object detection is essential for comprehensive scene understanding in autonomous driving. Despite its importance, LiDAR detection performance is limited by inherent constraints of point cloud data, particularly under conditions of extended distances and occlusions. Recently, temporal aggregation has been proven to significantly enhance detection accuracy by fusing multi-frame viewpoint information and enriching the spatial representation of objects. In this work, we introduce a novel LiDAR 3D object detection framework, namely LiSTM, to facilitate spatial-temporal feature learning with cross-frame motion forecasting information. We aim to improve the spatial-temporal interpretation capabilities of the LiDAR detector by incorporating a dynamic prior, generated from a non-learnable motion estimation model. Specifically, Motion-Guided Feature Aggregation (MGFA) is proposed to utilize the object trajectory from previous and future motion states to model spatial-temporal correlations into gaussian heatmap over a driving sequence. This motion-based heatmap then guides the temporal feature fusion, enriching the proposed object features. Moreover, we design a Dual Correlation Weighting Module (DCWM) that effectively facilitates the interaction between past and prospective frames through scene- and channel-wise feature abstraction. In the end, a cascade cross-attention-based decoder is employed to refine the 3D prediction. We have conducted experiments on the Waymo and nuScenes datasets to demonstrate that the proposed framework achieves superior 3D detection performance with effective spatial-temporal feature learning.

Reconstruction-based methods have significantly advanced modern unsupervised anomaly detection. However, the strong capacity of neural networks often violates the underlying assumptions by reconstructing abnormal samples well. To alleviate this issue, we present a simple yet effective reconstruction framework named Attention-Guided Pertuation Network (AGPNet), which learns to add perturbation noise with an attention mask, for accurate unsupervised anomaly detection. Specifically, it consists of two branches, \ie, a plain reconstruction branch and an auxiliary attention-based perturbation branch. The reconstruction branch is simply a plain reconstruction network that learns to reconstruct normal samples, while the auxiliary branch aims to produce attention masks to guide the noise perturbation process for normal samples from easy to hard. By doing so, we are expecting to synthesize hard yet more informative anomalies for training, which enable the reconstruction branch to learn important inherent normal patterns both comprehensively and efficiently. Extensive experiments are conducted on three popular benchmarks covering MVTec-AD, VisA, and MVTec-3D, and show that our framework obtains leading anomaly detection performance under various setups including few-shot, one-class, and multi-class setups.

People with visual impairments perceive their environment non-visually and often use AI-powered assistive tools to obtain textual descriptions of visual information. Recent large vision-language model-based AI-powered tools like Be My AI are more capable of understanding users' inquiries in natural language and describing the scene in audible text; however, the extent to which these tools are useful to visually impaired users is currently understudied. This paper aims to fill this gap. Our study with 14 visually impaired users reveals that they are adapting these tools organically -- not only can these tools facilitate complex interactions in household, spatial, and social contexts, but they also act as an extension of users' cognition, as if the cognition were distributed in the visual information. We also found that although the tools are currently not goal-oriented, users accommodate this limitation and embrace the tools' capabilities for broader use. These findings enable us to envision design implications for creating more goal-oriented, real-time processing, and reliable AI-powered assistive technology.

The Mean Square Error (MSE) is commonly utilized to estimate the solution of the optimal value function in the vast majority of offline reinforcement learning (RL) models and has achieved outstanding performance. However, we find that its principle can lead to overestimation phenomenon for the value function. In this paper, we first theoretically analyze overestimation phenomenon led by MSE and provide the theoretical upper bound of the overestimated error. Furthermore, to address it, we propose a novel Bellman underestimated operator to counteract overestimation phenomenon and then prove its contraction characteristics. At last, we propose the offline RL algorithm based on underestimated operator and diffusion policy model. Extensive experimental results on D4RL tasks show that our method can outperform state-of-the-art offline RL algorithms, which demonstrates that our theoretical analysis and underestimation way are effective for offline RL tasks.

Medical anomaly detection is a critical research area aimed at recognizing abnormal images to aid in diagnosis.Most existing methods adopt synthetic anomalies and image restoration on normal samples to detect anomaly. The unlabeled data consisting of both normal and abnormal data is not well explored. We introduce a novel Spatial-aware Attention Generative Adversarial Network (SAGAN) for one-class semi-supervised generation of health images.Our core insight is the utilization of position encoding and attention to accurately focus on restoring abnormal regions and preserving normal regions. To fully utilize the unlabelled data, SAGAN relaxes the cyclic consistency requirement of the existing unpaired image-to-image conversion methods, and generates high-quality health images corresponding to unlabeled data, guided by the reconstruction of normal images and restoration of pseudo-anomaly images.Subsequently, the discrepancy between the generated healthy image and the original image is utilized as an anomaly score.Extensive experiments on three medical datasets demonstrate that the proposed SAGAN outperforms the state-of-the-art methods., Comment: Early Accept by MICCAI 2024

The task of single-source domain generalization (SDG) in medical image segmentation is crucial due to frequent domain shifts in clinical image datasets. To address the challenge of poor generalization across different domains, we introduce a Plug-and-Play module for data augmentation called MoreStyle. MoreStyle diversifies image styles by relaxing low-frequency constraints in Fourier space, guiding the image reconstruction network. With the help of adversarial learning, MoreStyle further expands the style range and pinpoints the most intricate style combinations within latent features. To handle significant style variations, we introduce an uncertainty-weighted loss. This loss emphasizes hard-to-classify pixels resulting only from style shifts while mitigating true hard-to-classify pixels in both MoreStyle-generated and original images. Extensive experiments on two widely used benchmarks demonstrate that the proposed MoreStyle effectively helps to achieve good domain generalization ability, and has the potential to further boost the performance of some state-of-the-art SDG methods. Source code is available at https://github.com/zhaohaoyu376/morestyle., Comment: MICCAI2024

In clinical examinations and diagnoses, low-dose computed tomography (LDCT) is crucial for minimizing health risks compared with normal-dose computed tomography (NDCT). However, reducing the radiation dose compromises the signal-to-noise ratio, leading to degraded quality of CT images. To address this, we analyze LDCT denoising task based on experimental results from the frequency perspective, and then introduce a novel self-supervised CT image denoising method called WIA-LD2ND, only using NDCT data. The proposed WIA-LD2ND comprises two modules: Wavelet-based Image Alignment (WIA) and Frequency-Aware Multi-scale Loss (FAM). First, WIA is introduced to align NDCT with LDCT by mainly adding noise to the high-frequency components, which is the main difference between LDCT and NDCT. Second, to better capture high-frequency components and detailed information, Frequency-Aware Multi-scale Loss (FAM) is proposed by effectively utilizing multi-scale feature space. Extensive experiments on two public LDCT denoising datasets demonstrate that our WIA-LD2ND, only uses NDCT, outperforms existing several state-of-the-art weakly-supervised and self-supervised methods. Source code is available at https://github.com/zhaohaoyu376/WI-LD2ND., Comment: MICCAI2024

We present a comprehensive study of the light-cone generalized parton distribution (GPD) and quasi-GPD of a flavor-neutral meson in the 't Hooft model, {\it i.e.}, two-dimensional QCD (\QCDtw) in the $N_c\to\infty$ limit. With the aid of the Hamiltonian approach, we construct the light-cone GPD in terms of the meson's light-cone wave function in the framework of light-front quantization, and express the quasi-GPD in terms of the meson's Bars-Green wave functions and the chiral angle in the framework of equal-time quantization. We show that, both analytically and numerically, the quasi-GPD does approach the light-cone GPD when the meson is boosted to the infinite momentum frame, which justifies the tenet underlying the large momentum effective theory for the off-forward parton distribution. Upon taking the forward limit, the light-cone and quasi-GPDs reduce to the light-cone and quasi-PDFs. As a bonus, we take this chance to correct the incomplete expression of the quasi-PDFs in the 't Hooft model reported in our preceding work [Y. Jia et al. Phys. Rev. D 98, 054011 (2018)]., Comment: 31 pages, 14 figures, 1 table

In this paper, we build on our previous work to further investigate the role of evanescent operators in gauge theories, with a particular focus on their contribution to violations of unitarity. We develop an efficient method for calculating the norms of gauge-invariant operators in Yang-Mills (YM) theory by employing on-shell form factors. Our analysis, applicable to general spacetime dimensions, reveals the existence of negative norm states among evanescent operators. We also explore the one-loop anomalous dimensions of these operators and find complex anomalous dimensions. We broaden our analysis by considering YM theory coupled with scalar fields and we observe similar patterns of non-unitarity. The presence of negative norm states and complex anomalous dimensions across these analyses provides compelling evidence that general gauge theories are non-unitary in non-integer spacetime dimensions., Comment: 41 pages, 2 figures

In deep learning, stochastic gradient descent (SGD) and its momentum-based variants are widely used for optimization, but they typically suffer from slow convergence. Conversely, existing adaptive learning rate optimizers speed up convergence but often compromise generalization. To resolve this issue, we propose a novel optimization method designed to accelerate SGD's convergence without sacrificing generalization. Our approach reduces the variance of the historical gradient, improves first-order moment estimation of SGD by applying Wiener filter theory, and introduces a time-varying adaptive gain. Empirical results demonstrate that SGDF (SGD with Filter) effectively balances convergence and generalization compared to state-of-the-art optimizers.

Revealing the role of Coulomb interaction in topological semimetals with Dirac/Weyl-like band dispersion shapes a new frontier in condensed matter physics. Topological node-line semimetals (TNLSMs), anticipated as a fertile ground for exploring electronic correlation effects due to the anisotropy associated with their node-line structure, have recently attracted considerable attention. In this study, we report an experimental observation for correlation effects in TNLSMs realized by black phosphorus (BP) under hydrostatic pressure. By performing a combination of nuclear magnetic resonance measurements and band calculations on compressed BP, a magnetic-field-induced electronic instability of Weyl-like fermions is identified under an external magnetic field parallel to the so-called nodal ring in the reciprocal space. Anomalous spin fluctuations serving as the fingerprint of electronic instability are observed at low temperatures, and they are observed to maximize at approximately 1.0 GPa. This study presents compressed BP as a realistic material platform for exploring the rich physics in strongly coupled Weyl-like fermions., Comment: 10 pages, 4 figures

In various applications, such as robotic navigation and remote visual assistance, expanding the field of view (FOV) of the camera proves beneficial for enhancing environmental perception. Unlike image outpainting techniques aimed solely at generating aesthetically pleasing visuals, these applications demand an extended view that faithfully represents the scene. To achieve this, we formulate a new problem of faithful FOV extrapolation that utilizes a set of pre-captured images as prior knowledge of the scene. To address this problem, we present a simple yet effective solution called NeRF-Enhanced Outpainting (NEO) that uses extended-FOV images generated through NeRF to train a scene-specific image outpainting model. To assess the performance of NEO, we conduct comprehensive evaluations on three photorealistic datasets and one real-world dataset. Extensive experiments on the benchmark datasets showcase the robustness and potential of our method in addressing this challenge. We believe our work lays a strong foundation for future exploration within the research community.

Chiral symmetry plays an indispensable role in topological classifications as well as in the understanding of the origin of bulk or boundary flat bands. The conventional definition of chiral symmetry refers to the existence of a constant unitary matrix anticommuting with the Hamiltonian. As a constant unitary matrix has constant eigenvectors, boundary flat bands enforced by chiral symmetry, which share the same eigenvectors with the chiral symmetry operator, are dictated to carry fixed (pseudo)spin polarizations and be featureless in quantum geometry. In this work, we generalize the chiral symmetry and introduce a concept termed sub-chiral symmetry. Unlike the conventional chiral symmetry operator defined as constant matrix, the sub-chiral symmetry operator depends on partial components of the momentum vector, so as its eigenvectors. We show that topological gapped or gapless systems without chiral symmetry but with sub-chiral symmetry can support boundary flat bands, which exhibit topological spin textures and quantized Berry phases. We expect that such intriguing boundary flat bands could give rise to a variety of exotic physics in the presence of interactions or disorders., Comment: 6+6 pages, 2 figures

Three-dimensional (3D) two-band Hopf insulators are a paradigmatic example of topological phases beyond the topological classifications based on powerful methods like $K$-theory and symmetry indicators.Since this class of topological insulating phases was theoretically proposed in 2008, they have attracted significant interest owing to their conceptual novelty, connection to knot theory, and many fascinating physical properties. However, because their realization requires special forms of long-range spin-orbit coupling (SOC), they have not been achieved in any 3D system yet. Here we report the first experimental realization of the long-sought-after Hopf insulator in a 3D circuit system. To implement the Hopf insulator, we construct basic pseudo-spin modules and connection modules that can realize $2\times2$-matrix elements and then design the circuit network according to a tight-binding Hopf insulator Hamiltonian constructed by the Hopf map. By simulating the band structure of the designed circuit network and calculating the Hopf invariant, we find that the circuit realizes a Hopf insulator with Hopf invariant equaling $4$. Experimentally, we measure the band structure of a printed circuit board and find the observed properties of the bulk bands and topological surface states (TSS) are in good agreement with the theoretical predictions, verifying the bulk-boundary correspondence of the Hopf insulator. Our scheme brings the experimental study of Hopf insulators to reality and opens the door to the implementation of more unexplored topological phases beyond the known topological classifications.

We present concrete evidence that Yang-Mills theory exhibits non-unitarity in non-integer spacetime dimensions. This violation of unitarity stems from evanescent operators that, while vanishing in four dimensions, are non-zero in general d dimensions. We demonstrate that these evanescent operators lead to the emergence of both negative-norm states and complex anomalous dimensions., Comment: 6 pages