Tianhang Tang

SVPLP-SLAM： An RGB-D visual SLAM using point-line-plane feature fusion with supposed plane constraints

YANG Lei, Jie Chen, Tianhang Tang, LIU Yiguang

Traditional RGB-D Visual Simultaneous Localization and Mapping （vSLAM） systems rely predominantly on a single kind of feature or a simplistic combination of features.However， such approaches often fail to achieve optimal accuracy and mapping completeness， particularly in scenarios characterized by low texture and …

PGDIFF: Prior-Consistency Guided Diffusion for Unsupervised Image Restoration under Adverse Weather Conditions

Tianhang Tang, Jie Chen, Shun Lv, Ling Lei et autres

Image restoration under adverse weather conditions plays a crucial role in visual perception and downstream vision tasks. However, existing supervised methods heavily rely on paired data that is typically infeasible in practice, while unsupervised GAN-based approaches often suffer from training instability and …

cn (code pays fourni par la source)

Multi-Scale K-Cosine Curvature Fusion With Nonlinear Measure for Robust Corner Detection

Jie Chen, Tianhang Tang, Shun Lv, Yiguang Liu

cn (code pays fourni par la source)

Efficient guided diffusion toward adverse weather image restoration

Tianhang Tang, Jie Chen, Shun Lv, Ling Lei et autres

cn (code pays fourni par la source)

Efficient Guided Diffusion Toward Adverse Weather Image Restoration

Tianhang Tang, Jie Chen, Shun Lv, Ling Lei et autres

cn (code pays fourni par la source)

A Point-Line Features Fusion Method for Fast and Robust Monocular Visual-Inertial Initialization

Guoqiang Xie, Jie Chen, Tianhang Tang, Zeyu Chen et autres

Fast and robust initialization is essential for highly accurate monocular visual-inertial odometer (VIO), but at present majority of initialization methods rely only on point features, unstable in low texture and blurring situations. Therefore, we propose a novel point-line features fusion method for …

cn (code pays fourni par la source)

Enhancing in full-color single-pixel imaging: integrating variable density sampling with hyper-Laplacian priors

Shun Lv, Tianhang Tang, Jie Chen, Xuelei Shi et autres

Full-color single-pixel imaging aims to restore chromatic images using a single detector element, such as a photodiode or a single-pixel camera. However, image quality is inevitably compromised at low sampling rates due to inefficient sampling methods or incomplete representation of spectrum information. …

cn (code pays fourni par la source)

AST: An Attention-Guided Segment Transformer for Drone-Based Cross-View Geo-Localization

Zichuan Zhao, Tianhang Tang, Jie Chen, Xuelei Shi et autres

cn (code pays fourni par la source)

Precise Depth Estimation by Calculating Affine Transformation Parameters

Xuelei Shi, Tianhang Tang, Jie Chen, Shun Lv et autres

When estimating depth from Unmanned Aerial Vehicle (UAV) consecutive frames, usually only narrow baseline stereo is satisfied. So the estimation needs a higher image matching accuracy often not achieved in real applications. To tackle this problem, we propose a precise and robust …

cn (code pays fourni par la source)

Survival Analysis of the Titanic Using Random Forests

Tianhang Tang

cn (code pays fourni par la source)

An Integrated Platform for in vivo Electrophysiology in Spatial Cognition Experiments

Alfonso Brea Guerrero, M. Oijala, Shawn C. Moseley, Tianhang Tang et autres

Abstract Spatial cognition research requires behavioral paradigms that can distinguish between different navigational elements, such as allocentric (map-like) navigation and egocentric (e.g., body centered) navigation. To fill this need, we developed a flexible experimental platform that can be quickly modified without the …

us, ca (code pays fourni par la source)

Spatial temporal Fourier single-pixel imaging

Zixin Tang, Tianhang Tang, Jie Chen, Shun Lv et autres

Generally, the imaging quality of Fourier single-pixel imaging (FSI) will severely degrade while achieving high-speed imaging at a low sampling rate (SR). To tackle this problem, a new, to the best of our knowledge, imaging technique is proposed: firstly, the Hessian-based norm …

cn (code pays fourni par la source)