arXiv subjects

Aihong Tan

Publications and source records attributed to Aihong Tan.

9 recordsLinked to original sources

arXiv2025

ADA-DPM: A Neural Descriptors-based Adaptive Noise Filtering Strategy for SLAM

Lidar SLAM plays a significant role in mobile robot navigation and high-definition map construction. However, existing methods often face a trade-off between localization accuracy and system robustness in scenarios with a high proportion of dynamic objects, point cloud distortion, and unstructured environments. To address this issue, we propose a neural descriptors-based adaptive noise filtering strategy for SLAM, named ADA-DPM, which improves the performance of localization and mapping tasks through three key technical innovations. Firstly, to tackle dynamic object interference, we design the Dynamic Segmentation Head to predict and filter out dynamic feature points, eliminating the ego-motion interference caused by dynamic objects. Secondly, to mitigate the impact of noise and unstructured feature points, we propose the Global Importance Scoring Head that adaptively selects high-contribution feature points while suppressing the influence of noise and unstructured feature points. Moreover, we introduce the Cross-Layer Graph Convolution Module (GLI-GCN) to construct multi-scale neighborhood graphs, fusing local structural information across different scales and improving the discriminative power of overlapping features. Finally, experimental validations on multiple public datasets confirm the effectiveness of ADA-DPM.

cs.RO↗

arXiv2024

MS23D: A 3D Object Detection Method Using Multi-Scale Semantic Feature Points to Construct 3D Feature Layer

LiDAR point clouds can effectively depict the motion and posture of objects in three-dimensional space. Many studies accomplish the 3D object detection by voxelizing point clouds. However, in autonomous driving scenarios, the sparsity and hollowness of point clouds create some difficulties for voxel-based methods. The sparsity of point clouds makes it challenging to describe the geometric features of objects. The hollowness of point clouds poses difficulties for the aggregation of 3D features. We propose a two-stage 3D object detection framework, called MS23D. (1) We propose a method using voxel feature points from multi-branch to construct the 3D feature layer. Using voxel feature points from different branches, we construct a relatively compact 3D feature layer with rich semantic features. Additionally, we propose a distance-weighted sampling method, reducing the loss of foreground points caused by downsampling and allowing the 3D feature layer to retain more foreground points. (2) In response to the hollowness of point clouds, we predict the offsets between deep-level feature points and the object's centroid, making them as close as possible to the object's centroid. This enables the aggregation of these feature points with abundant semantic features. For feature points from shallow-level, we retain them on the object's surface to describe the geometric features of the object. To validate our approach, we evaluated its effectiveness on both the KITTI and ONCE datasets.

cs.CV↗

arXiv2024

PV-SSD: A Multi-Modal Point Cloud Feature Fusion Method for Projection Features and Variable Receptive Field Voxel Features

LiDAR-based 3D object detection and classification is crucial for autonomous driving. However, real-time inference from extremely sparse 3D data is a formidable challenge. To address this problem, a typical class of approaches transforms the point cloud cast into a regular data representation (voxels or projection maps). Then, it performs feature extraction with convolutional neural networks. However, such methods often result in a certain degree of information loss due to down-sampling or over-compression of feature information. This paper proposes a multi-modal point cloud feature fusion method for projection features and variable receptive field voxel features (PV-SSD) based on projection and variable voxelization to solve the information loss problem. We design a two-branch feature extraction structure with a 2D convolutional neural network to extract the point cloud's projection features in bird's-eye view to focus on the correlation between local features. A voxel feature extraction branch is used to extract local fine-grained features. Meanwhile, we propose a voxel feature extraction method with variable sensory fields to reduce the information loss of voxel branches due to downsampling. It avoids missing critical point information by selecting more useful feature points based on feature point weights for the detection task. In addition, we propose a multi-modal feature fusion module for point clouds. To validate the effectiveness of our method, we tested it on the KITTI dataset and ONCE dataset.

cs.CV↗

arXiv2010

Toward demonstrating controlled-X operation based on continuous variable four-partite cluster state and quantum teleporters

One-way quantum computation based on measurement and multipartite cluster entanglement offers the ability to perform a variety of unitary operations only through different choices of measurement bases. Here we present an experimental study toward demonstrating the controlled-X operation, a two-mode gate, in which continuous variable (CV) four-partite cluster states of optical modes are utilized. Two quantum teleportation elements are used for achieving the gate operation of the quantum state transformation from input target and control states to output states. By means of the optical cluster state prepared off-line, the homodyne detection and electronic feeding forward, the information carried by the input control state is transformed to the output target state. The presented scheme of the controlled-X operation based on teleportation can be implemented nonlocally and deterministically. The distortion of the quantum information resulting from the imperfect cluster entanglement is estimated with the fidelity.

quant-ph↗

arXiv2009

Quantum Logical Gates with Linear Quadripartite Cluster States of Continuous Variables

The concrete schemes to realize three types of basic quantum logical gates using linear quadripartite cluster states of optical continuous variables are proposed. The influences of noises and finite squeezing on the computation precision are analyzed in terms of the fidelity of propagated quantum information through the continuous variable cluster states. The proposed schemes provide direct references for the design of experimental systems of one-way quantum computer based on the cluster entanglement of amplitude and phase quadratures of light.

quant-ph↗

arXiv2008

Experimental generation of genuine four-partite entangled states with total three-party correlation for continuous variables

We experimentally prepare a new type of continuous variable genuine four-partite entangled states, the quantum correlation property of which is different from that of the four-mode GHZ and cluster states, and which has not any qubit counterpart to be proposed at present. In the criterion inequalities for the full inseparability of the genuine four-partite entangled states, the amplitude and phase quadrature correlation variances totally consisting of three-party combination from the four entangled modes are involved. The measured correlation variances among the quadratures of the prepared entangled states satisfy the sufficient requirements for the full inseparability. The type of entangled states has especially potential application in quantum information with continuous quantum variables.

quant-ph↗

arXiv2006

Experimental Preparation of Quadripartite Cluster and GHZ Entangled States for Continuous Variables

The cluster states and Greenberger-Horne-Zeilinger (GHZ) states are two different types of multipartite quantum entangled states. We present the first experimental results generating continuous variable quadripartite cluster and GHZ entangled states of electromagnetic fields. Utilizing four two-mode squeezed states of light and linearly optical transformations, the two types of entangled states for amplitude and phase quadratures of light are experimentally produced. The combinations of the measured quadrature variances prove the full inseparability of the generated four subsystems. The presented experimental schemes show that the multipartite entanglement of continuous variables can be deterministically generated with the relatively simple implementation.

quant-ph↗

arXiv2006

Frequency Conversion of Entangled State

The quantum characteristics of sum-frequency process in an optical cavity with an input signal optical beam, which is a half of entangled optical beams, are analyzed. The calculated results show that the quantum properties of the signal beam can be maintained after its frequency is conversed during the intracavity nonlinear optical interaction. The frequency-conversed output signal beam is still in an entangled state with the retained other half of initial entangled beams. The resultant quantum correlation spectra and the parametric dependences of the correlations on the initial squeezing factor, the optical losses and the pump power of the sum-frequency cavity are calculated. The proposed system for the frequency conversion of entangled state can be used in quantum communication network and the calculated results can provide direct references for the design of experimental systems.

quant-ph↗

arXiv2006

Experimental Demonstration of Quantum Entanglement Between Frequency-Nondegenerate Optical Twin Beams

The quantum entanglement of amplitude and phase quadratures between two intense optical beams with the total intensity of 22mW and the frequency difference of 1nm, which are produced from an optical parametric oscillator operating above threshold, is experimentally demonstrated with two sets of unbalanced Mach-Zehnder interferometers. The measured quantum correlations of intensity and phase are in reasonable agreement with the results calculated based on a semi-classical analysis of the noise characteristics given by C. Fabre et al.

quant-ph↗