Jun Liu

L’hypotension orthostatique chez les personnes atteintes de la maladie de Parkinson

Xue Zhu, Jun Liu, Andrei Krassioukov

L'hypotension orthostatique est souvent associée à la maladie de Parkinson, ce qui représente un risque important pour les personnes touchéesLa maladie de Parkinson touche entre 3,2 et 3,6 personnes sur 1000 au Canada.On estime que 36 % des personnes atteintes de la …

ca (code pays fourni par la source)

Sports-QA: A Large-Scale Video Question Answering Benchmark for Complex and Professional Sports

Haopeng Li, Andong Deng, Jun Liu, Hossein Rahmani et autres

Abstract Reasoning over sports videos for question answering is an important task with numerous applications, such as player training and information retrieval. However, this task has not been explored due to the lack of relevant datasets and the challenging nature it presents. …

SGDFuse: SAM-guided diffusion model for high-fidelity infrared and visible image fusion

XiaoYang Zhang, Jie Li, Guodong Fan, Yakun Ju et autres

cn, gb, sk (code pays fourni par la source)

SAM-Zero3D: Extending Segment Anything to Zero Shot 3D Scene Segmentation via Iterative Global–Local Interaction

Dejun Zhang, Shifeng Xu, Yanzi Bai, Yiqi Wu et autres

Lifting multi-view 2D masks generated by the Segment Anything Model (SAM) into 3D space offers a promising direction for zero-shot 3D scene segmentation, but view-dependent occlusions and limited fields of view often cause incomplete observations and cross-view inconsistencies, resulting in fragmented semantics …

cn, gb (code pays fourni par la source)

Pose-Guided Multi-Cue Explicit Query Construction for Disambiguating Human--Object Interactions

Minghao Zou, Shangkun Liu, Qingtian Zeng, Xue Zhang et autres

Human-Object Interaction (HOI) detection remains challenging due to the semantic ambiguity of interaction categories and the limited discriminability of their feature representations. Existing approaches often improve recognition by employing sophisticated models or auxiliary textual annotations. While effective in certain gains, these solutions …

cn, gb (code pays fourni par la source)

Unleashing the Power of Text-to-Image Diffusion Models for Category-Agnostic Pose Estimation

Duo Peng, Zhengbo Zhang, Ping Hu, Qiuhong Ke et autres

Category-Agnostic Pose Estimation (CAPE) aims to detect keypoints of unseen object categories in a few-shot setting, where the scarcity of labeled data poses significant challenges to generalization. In this work, we propose Prompt Pose Matching (PPM), a novel framework that unleashes the …

sg, cn, au, gb (code pays fourni par la source)

DifAttack++: Query-Efficient Black-Box Adversarial Attack via Hierarchical Disentangled Feature Space in Cross–Domain

Jun Liu, Jiantao Zhou, Jiandian Zeng, Jinyu Tian

This work investigates efficient score-based black-box adversarial attacks that achieve a high Attack Success Rate (ASR) and good generalization ability. We propose a novel attack framework, termed DifAttack++, which operates in a hierarchical disentangled feature space and significantly differs from existing methods …

mo, cn, jp (code pays fourni par la source)

PePNet: Pose-Enhanced Point Cloud Network for LiDAR-Based Human Action Recognition in Outdoor Long-Range Scenarios

Mengyuan Liu, Zhichao Deng, W. Zhang, Zi Wang et autres

With potential applications in robotics and autonomous vehicles, LiDAR-based human action recognition (HAR) in outdoor long-range scenarios is challenging due to the degradation of point cloud density with distance and the simultaneous motion of humans and sensors. To address these issues, we …

cn, gb (code pays fourni par la source)

Niagara : Normal-Integrated Geometric Affine Field for Scene Reconstruction From a Single View

Xianzu Wu, Zhenxin Ai, Chao Yang, Ser-Nam Lim et autres

Recent advances insingle-view3D scene reconstruction have highlighted the challenges in capturing fine geometric details and ensuring structural consistency, particularly in high-fidelity outdoor scene modeling. This paper presents Niagara, a new single-view 3D scene reconstruction framework that can faithfully reconstruct challenging outdoor scenes …

cn, hk, us, gb (code pays fourni par la source)

Whole-exome sequencing and burden analysis identify six novel candidate risk genes and expand the genetic landscape of Parkinson’s disease

Fan Yu, Zhen Hu, Qinqin Yan, Jun Liu

Parkinson's Disease (PD) is a complex neurodegenerative disorder with a largely undefined genetic architecture, particularly regarding the role of rare coding variants. We performed a large-scale exome-wide association study to systematically identify rare genetic risk factors for PD. We analyzed whole-exome sequencing …

cn (code pays fourni par la source)

MAIENet: Multi-Modality Adaptive Interaction Enhancement Network for SAR Object Detection

Tong Yu, Kai Xiong, Jun Liu, Guixing Cao et autres

Syntheticaperture radar (SAR) object detection offers significant advantages in remote sensing applications, particularly under adverse weather conditions or low-light environments. However, single-modal SAR image object detection encounters numerous challenges, including speckle noise, limited texture information, and interference from complex backgrounds. To address …

cn (code pays fourni par la source)

MantleMark: Migrating Watermarks from Multi-View Images to Radiance Fields via Frequency Modulation

Ziyuan Luo, Jun Liu, Haoliang Li, Renjie Wan

Multi-view images are essential for modern radiance field reconstruction methods like Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS). While image watermarking is a crucial data protection and ownership verification technique, it faces unprecedented challenges in multi-view scenarios. Traditional 2D watermarking …

hk, gb, br (code pays fourni par la source)