Lixiang Xu

Dr. Lixiang Xu


Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

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

Representative Publications

View All Publications
29 January 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
YCYan Chen
Yan Chen
QDQuan Dong
Quan Dong
XWXiaofeng Wang
Xiaofeng Wang
QZQianchuan Zhang
Qianchuan Zhang
MKMenglei Kang
Menglei Kang
CZChen Zhang
Chen Zhang
In the context of fast progress in deep learning, convolutional neural networks (CNNs) have been extensively applied to the semantic segmentation of remote sensing images and have achieved significant progress. However, certain limitations exist in capturing global contextual information due to the characteristics of convolutional local properties. Recently, Transformer has become a focus of research in computer vision and has shown great potential in extracting global contextual information, further promoting the development of semantic segmentation tasks. In this paper, we use ResNet50 as an encoder, embed the hybrid attention mechanism into Transformer, and propose a Transformer-based decoder. The Channel-Spatial Transformer Block (CSTB) further aggregates features by integrating the local feature maps extracted by the encoder with their associated global dependencies. At the same time, an adaptive approach is employed to reweight the interdependent channel maps to enhance the feature fusion. The Global Cross-Fusion Module (GCFM) combines the extracted complementary features to obtain more comprehensive semantic information. Extensive comparative experiments were conducted on the ISPRS Potsdam and Vaihingen datasets, where mIoU reached 78.06% and 76.37%, respectively. The outcomes of multiple ablation experiments also validate the effectiveness of the proposed method.

27 December 2024 in Pattern Recognition
Verified
Journal Article
Closed access
LXLixiang Xu
Lixiang Xu
JPJiawang Peng
Jiawang Peng
XJXiaoyi Jiang
Xiaoyi Jiang
ECEnhong Chen
Enhong Chen
BLBin Luo
Bin Luo

1 October 2020 in Pattern Recognition
Verified
Journal Article
Closed access
LXLixiang Xu
Lixiang Xu
XWXiaofeng Wang
Xiaofeng Wang
LBLu Bai
Lu Bai
JXJin Xiao
Jin Xiao
QLQi Liu
Qi Liu
BLBin Luo
Bin Luo

18 January 2025 in ACM Transactions on Information Systems
Verified
Journal Article
Open access
LXLixiang Xu
Lixiang Xu
YLYusheng Liu
Yusheng Liu
Tong XuTong Xu
Tong Xu
ECEnhong Chen
Enhong Chen
YTYuanyan Tang
Yuanyan Tang
The application of contrastive learning (CL) to collaborative filtering (CF) in recommender systems has achieved remarkable success. CL-based recommendation models mainly focus on creating multiple augmented views by employing different graph augmentation methods and utilizing these views for self-supervised learning. However, current CL methods for recommender systems usually struggle to fully address the problem of noisy data. To address this problem, we propose the G raph A ugmentation E mpowered C ontrastive L earning (GAECL) for recommendation framework, which uses graph augmentation based on topological and semantic dual adaptation and global co-modeling via structural optimization to co-create contrasting views for better augmentation of the CF paradigm. Specifically, we strictly filter out unimportant topologies by reconstructing the adjacency matrix and mask unimportant attributes in nodes according to the PageRank centrality principle to generate an augmented view that filters out noisy data. Additionally, GAECL achieves global collaborative modeling through structural optimization and generates another augmented view based on the PageRank centrality principle. This helps to filter the noisy data while preserving the original semantics of the data for more effective data augmentation. Extensive experiments are conducted on five datasets to demonstrate the superior performance of our model over various recommendation models.

22 March 2024 in IEEE Transactions on Learning Technologies
Verified
Journal Article
Closed access
LXLixiang Xu
Lixiang Xu
ZWZhanlong Wang
Zhanlong Wang
SZSuojuan Zhang
Suojuan Zhang
Xin YuanXin Yuan
Xin Yuan
MWMinjuan Wang
Minjuan Wang
ECEnhong Chen
Enhong Chen
Knowledge tracing (KT) is an intelligent educational technology used to model students’ learning progress and mastery in adaptive learning environments for personalized education. Despite utilizing deep learning models in KT, current approaches often oversimplify students’ exercise records into knowledge sequences, which fail to explore the rich information within individual questions. Additionally, existing KT models tend to neglect the complex, higher-order relationships between questions and latent concepts. Therefore, we introduce a novel model called Feature Crosses Information-based Knowledge Tracing (FCIKT) to explore the intricate interplay between questions, latent concepts, and question difficulties. FCIKT utilizes a fusion module to perform feature crosses operations on questions, integrating information from our constructed multi-relational heterogeneous graph using graph convolutional networks. We deployed a multi-head attention mechanism, which enriches the static embedding representations of questions and concepts with dynamic semantic information to simulate real-world scenarios of problem-solving. We also used gated recurrent units to dynamically capture and update the students’ knowledge state for final prediction. Extensive experiments demonstrated the validity and interpretability of our proposed model.

29 April 2024 in IEEE Transactions on Multimedia
Verified
Journal Article
Closed access
LXLixiang Xu
Lixiang Xu
QCQingzhe Cui
Qingzhe Cui
RHRichang Hong
Richang Hong
WXWei Xu
Wei Xu
ECEnhong Chen
Enhong Chen
YTYuan Yan Tang
Yuan Yan Tang
In recent years, the results of view-based 3D shape recognition methods have saturated, and models with excellent performance cannot be deployed on memory-limited devices due to their huge size of parameters. To address this problem, we introduce a compression method based on knowledge distillation for this field, which largely reduces the number of parameters while preserving model performance as much as possible. Specifically, to enhance the capabilities of smaller models, we design a high-performing large model called Group Multi-view Vision Transformer (GMViT). In GMViT, the view-level ViT first establishes relationships between view-level features. Additionally, to capture deeper features, we employ the grouping module to enhance view-level features into group-level features. Finally, the group-level ViT aggregates group-level features into complete, well-formed 3D shape descriptors. Notably, in both ViTs, we introduce spatial encoding of camera coordinates as innovative position embeddings. Furthermore, we propose two compressed versions based on GMViT, namely GMViT-simple and GMViT-mini. To enhance the training effectiveness of the small models, we introduce a knowledge distillation method throughout the GMViT process, where the key outputs of each GMViT component serve as distillation targets. Extensive experiments demonstrate the efficacy of the proposed method. The large model GMViT achieves excellent 3D classification and retrieval results on the benchmark datasets ModelNet, ShapeNetCore55, and MCB. The smaller models, GMViT-simple and GMViT-mini, reduce the parameter size by 8 and 17.6 times, respectively, and improve shape recognition speed by 1.5 times on average, while preserving at least 90% of the recognition performance. The code is available at https://github.com/bigdata-graph/GMViT .