arXiv subjects

Qi Wu

Publications and source records attributed to Qi Wu.

At least 163 records · Page 9Linked to original sources

arXiv2025

HyperFLINT: Hypernetwork-based Flow Estimation and Temporal Interpolation for Scientific Ensemble Visualization

We present HyperFLINT (Hypernetwork-based FLow estimation and temporal INTerpolation), a novel deep learning-based approach for estimating flow fields, temporally interpolating scalar fields, and facilitating parameter space exploration in spatio-temporal scientific ensemble data. This work addresses the critical need to explicitly incorporate ensemble parameters into the learning process, as traditional methods often neglect these, limiting their ability to adapt to diverse simulation settings and provide meaningful insights into the data dynamics. HyperFLINT introduces a hypernetwork to account for simulation parameters, enabling it to generate accurate interpolations and flow fields for each timestep by dynamically adapting to varying conditions, thereby outperforming existing parameter-agnostic approaches. The architecture features modular neural blocks with convolutional and deconvolutional layers, supported by a hypernetwork that generates weights for the main network, allowing the model to better capture intricate simulation dynamics. A series of experiments demonstrates HyperFLINT's significantly improved performance in flow field estimation and temporal interpolation, as well as its potential in enabling parameter space exploration, offering valuable insights into complex scientific ensembles.

cs.CV↗

arXiv2025

A Comprehensive Analysis of Mamba for 3D Volumetric Medical Image Segmentation

Mamba, with its selective State Space Models (SSMs), offers a more computationally efficient solution than Transformers for long-range dependency modeling. However, there is still a debate about its effectiveness in high-resolution 3D medical image segmentation. In this study, we present a comprehensive investigation into Mamba's capabilities in 3D medical image segmentation by tackling three pivotal questions: Can Mamba replace Transformers? Can it elevate multi-scale representation learning? Is complex scanning necessary to unlock its full potential? We evaluate Mamba's performance across three large public benchmarks-AMOS, TotalSegmentator, and BraTS. Our findings reveal that UlikeMamba, a U-shape Mamba-based network, consistently surpasses UlikeTrans, a U-shape Transformer-based network, particularly when enhanced with custom-designed 3D depthwise convolutions, boosting accuracy and computational efficiency. Further, our proposed multi-scale Mamba block demonstrates superior performance in capturing both fine-grained details and global context, especially in complex segmentation tasks, surpassing Transformer-based counterparts. We also critically assess complex scanning strategies, finding that simpler methods often suffice, while our Tri-scan approach delivers notable advantages in the most challenging scenarios. By integrating these advancements, we introduce a new network for 3D medical image segmentation, positioning Mamba as a transformative force that outperforms leading models such as nnUNet, CoTr, and U-Mamba, offering competitive accuracy with superior computational efficiency. This study provides key insights into Mamba's unique advantages, paving the way for more efficient and accurate approaches to 3D medical imaging.

cs.CV↗

arXiv2025

3DGUT: Enabling Distorted Cameras and Secondary Rays in Gaussian Splatting

3D Gaussian Splatting (3DGS) enables efficient reconstruction and high-fidelity real-time rendering of complex scenes on consumer hardware. However, due to its rasterization-based formulation, 3DGS is constrained to ideal pinhole cameras and lacks support for secondary lighting effects. Recent methods address these limitations by tracing the particles instead, but, this comes at the cost of significantly slower rendering. In this work, we propose 3D Gaussian Unscented Transform (3DGUT), replacing the EWA splatting formulation with the Unscented Transform that approximates the particles through sigma points, which can be projected exactly under any nonlinear projection function. This modification enables trivial support of distorted cameras with time dependent effects such as rolling shutter, while retaining the efficiency of rasterization. Additionally, we align our rendering formulation with that of tracing-based methods, enabling secondary ray tracing required to represent phenomena such as reflections and refraction within the same 3D representation. The source code is available at: https://github.com/nv-tlabs/3dgrut.

cs.GR↗

arXiv2025

EnvPoser: Environment-aware Realistic Human Motion Estimation from Sparse Observations with Uncertainty Modeling

Estimating full-body motion using the tracking signals of head and hands from VR devices holds great potential for various applications. However, the sparsity and unique distribution of observations present a significant challenge, resulting in an ill-posed problem with multiple feasible solutions (i.e., hypotheses). This amplifies uncertainty and ambiguity in full-body motion estimation, especially for the lower-body joints. Therefore, we propose a new method, EnvPoser, that employs a two-stage framework to perform full-body motion estimation using sparse tracking signals and pre-scanned environment from VR devices. EnvPoser models the multi-hypothesis nature of human motion through an uncertainty-aware estimation module in the first stage. In the second stage, we refine these multi-hypothesis estimates by integrating semantic and geometric environmental constraints, ensuring that the final motion estimation aligns realistically with both the environmental context and physical interactions. Qualitative and quantitative experiments on two public datasets demonstrate that our method achieves state-of-the-art performance, highlighting significant improvements in human motion estimation within motion-environment interaction scenarios.

cs.CV↗

arXiv2025

Generalized Tensor-Aided Channel Estimation for Hardware Impaired Device Identification

In this paper, we investigate the joint generalized channel estimation and device identification problem in Internet of Things (IoT) networks {under multipath propagation}. To fully utilize the received signal, we decompose the generalized channel into three components: transmitter hardware characteristics, path gains, and angles of arrival. By modelling the received signals as parallel factor (PARAFAC) tensors, we develop alternating least squares (ALS)-based algorithms to simultaneously estimate the generalized channels and identify the transmitters. Simulation results show that the proposed scheme outperforms {both Khatri-Rao Factorization (KRF) and the conventional least squares (LS) method} in terms of channel estimation accuracy and achieves performance close to the derived Cramer-Rao lower bound.

eess.SP↗

arXiv2025

MADS: Multi-Attribute Document Supervision for Zero-Shot Image Classification

Zero-shot learning (ZSL) aims to train a model on seen classes and recognize unseen classes by knowledge transfer through shared auxiliary information. Recent studies reveal that documents from encyclopedias provide helpful auxiliary information. However, existing methods align noisy documents, entangled in visual and non-visual descriptions, with image regions, yet solely depend on implicit learning. These models fail to filter non-visual noise reliably and incorrectly align non-visual words to image regions, which is harmful to knowledge transfer. In this work, we propose a novel multi-attribute document supervision framework to remove noises at both document collection and model learning stages. With the help of large language models, we introduce a novel prompt algorithm that automatically removes non-visual descriptions and enriches less-described documents in multiple attribute views. Our proposed model, MADS, extracts multi-view transferable knowledge with information decoupling and semantic interactions for semantic alignment at local and global levels. Besides, we introduce a model-agnostic focus loss to explicitly enhance attention to visually discriminative information during training, also improving existing methods without additional parameters. With comparable computation costs, MADS consistently outperforms the SOTA by 7.2% and 8.2% on average in three benchmarks for document-based ZSL and GZSL settings, respectively. Moreover, we qualitatively offer interpretable predictions from multiple attribute views.

cs.CV↗

arXiv2025

Are Large Vision Language Models Good Game Players?

Large Vision Language Models (LVLMs) have demonstrated remarkable abilities in understanding and reasoning about both visual and textual information. However, existing evaluation methods for LVLMs, primarily based on benchmarks like Visual Question Answering and image captioning, often fail to capture the full scope of LVLMs' capabilities. These benchmarks are limited by issues such as inadequate assessment of detailed visual perception, data contamination, and a lack of focus on multi-turn reasoning. To address these challenges, we propose \method{}, a game-based evaluation framework designed to provide a comprehensive assessment of LVLMs' cognitive and reasoning skills in structured environments. \method{} uses a set of games to evaluate LVLMs on four core tasks: Perceiving, Question Answering, Rule Following, and End-to-End Playing, with each target task designed to assess specific abilities, including visual perception, reasoning, decision-making, etc. Based on this framework, we conduct extensive experiments that explore the limitations of current LVLMs, such as handling long structured outputs and perceiving detailed and dense elements. Code and data are publicly available at https://github.com/xinke-wang/LVLM-Playground.

cs.CV↗

arXiv2025

mmDEAR: mmWave Point Cloud Density Enhancement for Accurate Human Body Reconstruction

Millimeter-wave (mmWave) radar offers robust sensing capabilities in diverse environments, making it a highly promising solution for human body reconstruction due to its privacy-friendly and non-intrusive nature. However, the significant sparsity of mmWave point clouds limits the estimation accuracy. To overcome this challenge, we propose a two-stage deep learning framework that enhances mmWave point clouds and improves human body reconstruction accuracy. Our method includes a mmWave point cloud enhancement module that densifies the raw data by leveraging temporal features and a multi-stage completion network, followed by a 2D-3D fusion module that extracts both 2D and 3D motion features to refine SMPL parameters. The mmWave point cloud enhancement module learns the detailed shape and posture information from 2D human masks in single-view images. However, image-based supervision is involved only during the training phase, and the inference relies solely on sparse point clouds to maintain privacy. Experiments on multiple datasets demonstrate that our approach outperforms state-of-the-art methods, with the enhanced point clouds further improving performance when integrated into existing models.

cs.CV↗

arXiv2025

Radiative decays of $X(3872)$ in $D{\bar D}^*$ molecule scenario

We investigate the radiative decays of the $X(3872)$ to $γV~(V=ρ^0,\, ω)$ in the molecule scenario, where the $X(3872)$ is regarded as a pure hadronic molecule of the $D\bar{D}^*+c.c$ in an $S$-wave with the quantum numbers $J^{PC}=1^{++}$. The radiative processes were assumed to occur via the triangle hadronic loops, and the relevant calculations were conducted using an effective Lagrangian approach. It is found that the absolute decay widths are model-dependent, but the relative width ratio is rather independent of the model parameter. Moreover, the calculated results indicate that the radiative decays of the $X(3872)$ are strongly influenced by the molecular configuration characterized by the proportion of the charged and neutral constituents. We hope that the present calculations could be tested by the experimental measurements.

hep-ph↗

arXiv2025

Ground-level Viewpoint Vision-and-Language Navigation in Continuous Environments

Vision-and-Language Navigation (VLN) empowers agents to associate time-sequenced visual observations with corresponding instructions to make sequential decisions. However, generalization remains a persistent challenge, particularly when dealing with visually diverse scenes or transitioning from simulated environments to real-world deployment. In this paper, we address the mismatch between human-centric instructions and quadruped robots with a low-height field of view, proposing a Ground-level Viewpoint Navigation (GVNav) approach to mitigate this issue. This work represents the first attempt to highlight the generalization gap in VLN across varying heights of visual observation in realistic robot deployments. Our approach leverages weighted historical observations as enriched spatiotemporal contexts for instruction following, effectively managing feature collisions within cells by assigning appropriate weights to identical features across different viewpoints. This enables low-height robots to overcome challenges such as visual obstructions and perceptual mismatches. Additionally, we transfer the connectivity graph from the HM3D and Gibson datasets as an extra resource to enhance spatial priors and a more comprehensive representation of real-world scenarios, leading to improved performance and generalizability of the waypoint predictor in real-world environments. Extensive experiments demonstrate that our Ground-level Viewpoint Navigation (GVnav) approach significantly improves performance in both simulated environments and real-world deployments with quadruped robots.

cs.RO↗

arXiv2025

Secure and Efficient Watermarking for Latent Diffusion Models in Model Distribution Scenarios

Latent diffusion models have exhibited considerable potential in generative tasks. Watermarking is considered to be an alternative to safeguard the copyright of generative models and prevent their misuse. However, in the context of model distribution scenarios, the accessibility of models to large scale of model users brings new challenges to the security, efficiency and robustness of existing watermark solutions. To address these issues, we propose a secure and efficient watermarking solution. A new security mechanism is designed to prevent watermark leakage and watermark escape, which considers watermark randomness and watermark-model association as two constraints for mandatory watermark injection. To reduce the time cost of training the security module, watermark injection and the security mechanism are decoupled, ensuring that fine-tuning VAE only accomplishes the security mechanism without the burden of learning watermark patterns. A watermark distribution-based verification strategy is proposed to enhance the robustness against diverse attacks in the model distribution scenarios. Experimental results prove that our watermarking consistently outperforms existing six baselines on effectiveness and robustness against ten image processing attacks and adversarial attacks, while enhancing security in the distribution scenarios.

cs.CR↗

arXiv2025

Probabilistic Learning of Multivariate Time Series with Temporal Irregularity

Probabilistic forecasting of multivariate time series is essential for various downstream tasks. Most existing approaches rely on the sequences being uniformly spaced and aligned across all variables. However, real-world multivariate time series often suffer from temporal irregularities, including nonuniform intervals and misaligned variables, which pose significant challenges for accurate forecasting. To address these challenges, we propose an end-to-end framework that models temporal irregularities while capturing the joint distribution of variables at arbitrary continuous-time points. Specifically, we introduce a dynamic conditional continuous normalizing flow to model data distributions in a non-parametric manner, accommodating the complex, non-Gaussian characteristics commonly found in real-world datasets. Then, by leveraging a carefully factorized log-likelihood objective, our approach captures both temporal and cross-sectional dependencies efficiently. Extensive experiments on a range of real-world datasets demonstrate the superiority and adaptability of our method compared to existing approaches.

cs.LG↗

arXiv2025

Open-Nav: Exploring Zero-Shot Vision-and-Language Navigation in Continuous Environment with Open-Source LLMs

Vision-and-Language Navigation (VLN) tasks require an agent to follow textual instructions to navigate through 3D environments. Traditional approaches use supervised learning methods, relying heavily on domain-specific datasets to train VLN models. Recent methods try to utilize closed-source large language models (LLMs) like GPT-4 to solve VLN tasks in zero-shot manners, but face challenges related to expensive token costs and potential data breaches in real-world applications. In this work, we introduce Open-Nav, a novel study that explores open-source LLMs for zero-shot VLN in the continuous environment. Open-Nav employs a spatial-temporal chain-of-thought (CoT) reasoning approach to break down tasks into instruction comprehension, progress estimation, and decision-making. It enhances scene perceptions with fine-grained object and spatial knowledge to improve LLM's reasoning in navigation. Our extensive experiments in both simulated and real-world environments demonstrate that Open-Nav achieves competitive performance compared to using closed-source LLMs.

cs.RO↗

arXiv2025

Exploring the nature of $Y(4230)$ and $Y(4360)$ in B decays

Vector charmonium states can be directly produced in the $e^+e^{-}$ annihilation process. Among them, $Y(4230)$ and $Y(4360)$ splitting from the previously discovered $Y(4260)$ are not easily arranged into the conventional charmonium spectrum, while recent studies have indicated that they have strong couplings to $D\bar{D}_1$ and $D^*\bar{D}_1$. In this work, we investigate the production of $Y(4230)$ and $Y(4360)$ as the heavy-quark spin symmetry doublet hadronic molecules of $D\bar{D}_1$ and $D^*\bar{D}_1$ in $B$ decays via the triangle diagram mechanism. In particular, we propose that the decay constants of $Y(4230)$ and $Y(4360)$ extracted in $B$ decays are useful for clarifying their nature.

hep-ph↗

arXiv2025

Superconducting-insulating phase transition in pressurized Ba1-xKxBiO3

We report the first observation of a pressure-induced transition from a superconducting (SC) to an insulating (I) phase in single-crystal Ba1-xKxBiO3 (x = 0.4, 0.43, 0.52, and 0.58) superconductors. X-ray diffraction measurements conducted at 20 K reveal a direct relationship between this SC-I transition and a pressure-induced distortion of crystal structure. With increasing pressure, the lattice parameters a and c of the ambient-pressure superconducting tetragonal (T) phase are compressed continuously below a critical pressure (Pc1), wherein the pressure (P) dependence of superconducting transition temperature (Tc) displays a small variation. However, upon further compression, the lattice of the compressed T phase displays an anisotropic change, and Tc shows a monotonous decrease. When the pressure reaches Pc2 (Pc2 > Pc1), the compressed T phase collapses along the c axis, followed by the disappearance of superconductivity and the appearance of the insulating phase. This SC-I transition is fully reversible, with the critical pressure increasing alongside K doping concentration. These findings are strikingly similar to the SC-I transition observed in hole-doped high-Tc cuprate superconductors under pressure. Identifying their commonalities could deepen our understanding of the mechanisms that underlie high- Tc superconductivity in these two oxide superconductors with a perovskite structure.

cond-mat.supr-con↗

arXiv2025

Investigations of key issues on the reproducibility of high-Tc superconductivity emerging from compressed La3Ni2O7

Recently, the signatures of superconductivity near 80 K have been discovered in the single crystal of La3Ni2O7 under pressure, which makes it a new candidate of the high-temperature superconductors dominated by 3d transition elements after the cuprate and iron-pnictide superconductors. However, there are several critical questions that have been perplexing the scientific community. These questions include (1) what factors contribute to the inconsistent reproducibility of the experimental results? (2) what is the fundamental nature of pressure-induced superconductivity: bulk or non-bulk (filamentary-like)? (3) where does the superconducting phase locate within the sample if it is filamentary-like? (4) is the oxygen content important for developing and stabilizing its superconductivity? In this study, we employ comprehensive high-pressure techniques to address these crucial issues. Through our modulated ac susceptibility measurements, we are the first to find that the superconductivity in this nickelate is filamentary-like. Our scanning transmission electron microscopy (STEM) investigations suggest that the filamentary-like superconductivity most likely emerges at the interface between the La3Ni2O7 and La4Ni3O10 phases. By tuning the oxygen content on the polycrystalline La3Ni2O7, we also find that the oxygen content plays vital role for developing and stabilizing its superconductivity. The upper and lower bounds of the oxygen content are 7.35 and 6.89, respectively. Our results provide not only new insights into understanding the puzzling issues in this material, but also significant information for achieving a better understanding on the superconductivity of this material.

cond-mat.supr-con↗

arXiv2025

UD-Mamba: A pixel-level uncertainty-driven Mamba model for medical image segmentation

Recent advancements have highlighted the Mamba framework, a state-space model known for its efficiency in capturing long-range dependencies with linear computational complexity. While Mamba has shown competitive performance in medical image segmentation, it encounters difficulties in modeling local features due to the sporadic nature of traditional location-based scanning methods and the complex, ambiguous boundaries often present in medical images. To overcome these challenges, we propose Uncertainty-Driven Mamba (UD-Mamba), which redefines the pixel-order scanning process by incorporating channel uncertainty into the scanning mechanism. UD-Mamba introduces two key scanning techniques: 1) sequential scanning, which prioritizes regions with high uncertainty by scanning in a row-by-row fashion, and 2) skip scanning, which processes columns vertically, moving from high-to-low or low-to-high uncertainty at fixed intervals. Sequential scanning efficiently clusters high-uncertainty regions, such as boundaries and foreground objects, to improve segmentation precision, while skip scanning enhances the interaction between background and foreground regions, allowing for timely integration of background information to support more accurate foreground inference. Recognizing the advantages of scanning from certain to uncertain areas, we introduce four learnable parameters to balance the importance of features extracted from different scanning methods. Additionally, a cosine consistency loss is employed to mitigate the drawbacks of transitioning between uncertain and certain regions during the scanning process. Our method demonstrates robust segmentation performance, validated across three distinct medical imaging datasets involving pathology, dermatological lesions, and cardiac tasks.

eess.IV↗

arXiv2025

General Scene Adaptation for Vision-and-Language Navigation

Vision-and-Language Navigation (VLN) tasks mainly evaluate agents based on one-time execution of individual instructions across multiple environments, aiming to develop agents capable of functioning in any environment in a zero-shot manner. However, real-world navigation robots often operate in persistent environments with relatively consistent physical layouts, visual observations, and language styles from instructors. Such a gap in the task setting presents an opportunity to improve VLN agents by incorporating continuous adaptation to specific environments. To better reflect these real-world conditions, we introduce GSA-VLN, a novel task requiring agents to execute navigation instructions within a specific scene and simultaneously adapt to it for improved performance over time. To evaluate the proposed task, one has to address two challenges in existing VLN datasets: the lack of OOD data, and the limited number and style diversity of instructions for each scene. Therefore, we propose a new dataset, GSA-R2R, which significantly expands the diversity and quantity of environments and instructions for the R2R dataset to evaluate agent adaptability in both ID and OOD contexts. Furthermore, we design a three-stage instruction orchestration pipeline that leverages LLMs to refine speaker-generated instructions and apply role-playing techniques to rephrase instructions into different speaking styles. This is motivated by the observation that each individual user often has consistent signatures or preferences in their instructions. We conducted extensive experiments on GSA-R2R to thoroughly evaluate our dataset and benchmark various methods. Based on our findings, we propose a novel method, GR-DUET, which incorporates memory-based navigation graphs with an environment-specific training strategy, achieving state-of-the-art results on all GSA-R2R splits.

cs.CV↗