Guoqiang Xiao

Prof. Dr. Guoqiang Xiao

orcid

Research Focus & Metrics

Academic Career Length

18Years (2009-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 August 2022 in IEEE Transactions on Multimedia
Verified
Journal Article
Closed access
XHXianfeng Han
Xianfeng Han
YJYi-Fei Jin
Yi-Fei Jin
HCHui-Xian Cheng
Hui-Xian Cheng
GXGuoqiang Xiao
Guoqiang Xiao
Feature representation learning is a key component in 3D point cloud analysis. However, the powerful convolutional neural networks (CNNs) cannot be applied due to the irregular structure of point clouds. Therefore, following the tremendous success of transformer in natural language processing and image understanding tasks, in this paper, we present a novel point cloud representation learning architecture, named Dual Transformer Network (DTNet), which mainly consists of Dual Point Cloud Transformer (DPCT) module. Specifically, by aggregating the well-designed point-wise and channel-wise self-attention models simultaneously, DPCT module can capture much richer contextual dependencies semantically from the perspective of position and channel. With the DPCT model as a fundamental component, we construct the DTNet for performing 3D point cloud analysis in an end-to-end manner. Extensive quantitative and qualitative experiments on publicly available benchmarks demonstrate the effectiveness of our transformer framework for the tasks of 3D point cloud classification, segmentation and visual object affordance understanding, achieving highly competitive performance in comparison with the state-of-the-art approaches. The source code and trained models are available at https://github.com/hxf-666/DT.

28 February 2023 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
HCHui-Xian Cheng
Hui-Xian Cheng
XHXianfeng Han
Xianfeng Han
GXGuoqiang Xiao
Guoqiang Xiao
Effective and efficient 3D semantic segmentation from large-scale LiDAR point cloud is a fundamental problem in the field of autonomous driving. In this paper, we present Transformer-Range-View Network (TransRVNet), a novel and powerful projection-based CNN-Transformer architecture to infer point-wise semantics. First, a Multi Residual Channel Interaction Attention Module (MRCIAM) is introduced to capture channel-level multi-scale feature and model intra-channel, inter-channel correlations based on attention mechanism. Then, in the encoder stage, we use a well-designed Residual Context Aggregation Module (RCAM), including a residual dilated convolution structure and a context aggregation module, to fuse information from different receptive fields while reducing the impact of missing points. Finally, a Balanced Non-square-Transformer Module (BNTM) is employed as fundamental component of decoder to achieve locally feature dependencies for more discriminative feature learning by introducing the non-square shifted window strategy. Extensive qualitative and quantitative experiments conducted on challenging SemanticKITTI and SemanticPOSS benchmarks have verified the effectiveness of our proposed technique. Our TransRVNet presents superior performance over most existing state-of-the-art approaches. The source code and trained model are available at https://github.com/huixiancheng/TransRVNet.

1 February 2022 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
XHXianfeng Han
Xianfeng Han
ZHZhang-Yue He
Zhang-Yue He
JCJia Chen
Jia Chen
GXGuoqiang Xiao
Guoqiang Xiao
Self-attention mechanism recently achieves impressive advancement in Natural Language Processing (NLP) and Image Processing domains. Its permutation invariance property makes it ideally suitable for point cloud processing. Inspired by this remarkable success, we propose an end-to-end architecture, dubbed Cross-Level Cross-Scale Cross-Attention Network (3CROSSNet), for point cloud representation learning. First, a point-wise feature pyramid module is introduced to hierarchically extract features from different scales or resolutions. Then a cross-level cross-attention module is designed to model long-range inter-level and intra-level dependencies. Finally, we develop a cross-scale cross-attention module to capture interactions between-and-within scales for representation enhancement. Compared with state-of-the-art approaches, our network can obtain competitive performance on challenging 3D object classification, point cloud segmentation tasks via comprehensive experimental evaluation. The source code and trained models are available at. 1

12 April 2023 in Artificial Intelligence Review
Verified
Journal Article
Closed access
PDF file
XHXianfeng Han
Xianfeng Han
ZFZhi-Ao Feng
Zhi-Ao Feng
SSShi-Jie Sun
Shi-Jie Sun
GXGuoqiang Xiao
Guoqiang Xiao

15 June 2018 in IEEE Transactions on Signal Processing
Verified
Journal Article
Closed access
HXHanshen Xiao
Hanshen Xiao
YHYufeng Huang
Yufeng Huang
YYYu Ye
Yu Ye
GXGuoqiang Xiao
Guoqiang Xiao
Chinese Remainder Theorem (CRT) has been widely studied with its applications in frequency estimation, phase unwrapping, coding theory and distributed data storage. Since traditional CRT is greatly sensitive to the errors in residues due to noises, the problem of robustly reconstructing integers via the erroneous residues has been intensively studied in the literature. In order to robustly reconstruct integers, there are basically two approaches: one is to introduce common divisors in the moduli and the other is to directly decrease the dynamic range. In this paper, we take further insight into the geometry property of the linear space associated with CRT. Echoing both ways to introduce redundancy, we propose a pseudo metric as a uniform framework to analyze the trade-off between the error bound and the dynamic range for robust CRT. Furthermore, we present the first robust CRT for multiple numbers to solve the problem raised by CRT- based undersampling frequency estimation in general. Based on symmetric polynomials proposed, we proved that in most cases, the problem can be solved in polynomial time efficiently.

1 April 2016 in ETRI Journal
Verified
Journal Article
Open access
PDF file
HXHanshen Xiao
Hanshen Xiao
HGHari Krishna Garg
Hari Krishna Garg
JHJianhao Hu
Jianhao Hu
GXGuoqiang Xiao
Guoqiang Xiao
We propose and describe new error control algorithms for redundant residue number systems (RRNSs) and residue number system product codes. These algorithms employ search techniques for obtaining error values from within a set of values (that contains all possible error values). For a given RRNS, the error control algorithms have a computational complexity of t·O(log2 n + log2 mֿ) comparison operations, where t denotes the error correcting capability, n denotes the number of moduli, and mֿ denotes the geometric average of moduli. These algorithms avoid most modular operations. We describe a refinement to the proposed algorithms that further avoids the modular operation required in their respective first steps, with an increase of ⌈log2 n⌉ to their computational complexity. The new algorithms provide significant computational advantages over existing methods.