Low-overlap point cloud registration network integrating position encoding and overlap masks

doi:10.11772/j.issn.1001-9081.2024121782

Journal of Computer Applications ›› 2026, Vol. 46 ›› Issue (2): 536-545.DOI: 10.11772/j.issn.1001-9081.2024121782

• Multimedia computing and computer simulation • Previous Articles

Low-overlap point cloud registration network integrating position encoding and overlap masks

Xiaowei LA¹, Lihua HU¹(), Jianhua HU², Xiaoling YAO¹, Xinbo WANG²

^1.College of Computer Science and Technology，Taiyuan University of Science and Technology，Taiyuan Shanxi 030024，China
^2.Institute of Automation，Chinese Academy of Sciences，Beijing 100190，China

Received:2024-12-18 Revised:2025-05-14 Accepted:2025-05-15 Online:2025-06-05 Published:2026-02-10
Contact: Lihua HU
About author:LA Xiaowei， born in 2000， M. S. candidate. His research interests include computer vision， point cloud registration.
HU Lihua， born in 1982， Ph. D.， professor. Her research interests include computer vision， artificial intelligence， pattern recognition. Email:hlh@tyust.edu.cn
HU Jianhua， born in 1987， Ph. D.， associate research fellow. His research interests include intelligent robots，3D vision.
YAO Xiaoling， born in 1993， Ph. D. candidate. Her research interests include computer vision， data mining.
WANG Xinbo， born in 1987， Ph. D.， associate research fellow. His research interests include 3D vision， robots.
Supported by:
National Natural Science Foundation of China(62273248);Science and Technology Service Network Program of the Chinese Academy of Sciences(STS-HP-202202);Graduate Education Innovation Project of Taiyuan University of Science and Technology(BY2022017)

融合位置编码和重叠掩模的低重叠点云配准网络

喇孝伟¹, 胡立华¹(), 胡建华², 姚晓玲¹, 王欣波²

^1.太原科技大学计算机科学与技术学院，太原 030024
^2.中国科学院自动化研究所，北京 100190

通讯作者: 胡立华
作者简介:喇孝伟（2000—），男，河北张家口人，硕士研究生，主要研究方向：计算机视觉、点云配准
胡立华（1982—），女，山西忻州人，教授，博士，CCF会员，主要研究方向：计算机视觉、人工智能、模式识别 Email:hlh@tyust.edu.cn
胡建华（1987—），男，山西忻州人，副研究员，博士，主要研究方向：智能机器人、3D视觉
姚晓玲（1993—），女，山西忻州人，博士研究生，主要研究方向：计算机视觉、数据挖掘
王欣波（1987—），男，浙江嵊州人，副研究员，博士，主要研究方向：3D视觉、机器人。
基金资助:
国家自然科学基金资助项目(62273248);中国科学院科技服务网络计划项目(STS-HP-202202);太原科技大学研究生教育创新项目(BY2022017)

Abstract

Abstract:

For the issues of low registration accuracy and high mismatch rate in low-overlap point cloud registration due to insufficient descriptive information of keypoint features and minimal overlapping regions， this paper proposed a low-overlap point cloud registration network that integrates position encoding and overlap masks was proposed to reduce mismatch rate and improve registration accuracy. First， a PointNet-based point-wise feature encoder was employed to extract keypoints， which were then enhanced by fusing their feature information， coordinate data， and position encoding to generate more discriminative keypoint descriptors. Second， the fused features were processed through self-attention and cross-attention modules to strengthen the descriptive power of point cloud features and enhance contextual interaction， thereby addressing the problem of insufficient keypoint descriptive information. Third， an overlap mask module was introduced after the attention modules to filter out keypoints from non-overlapping regions through learned masks， further reducing mismatch rate. Finally， optimal matching was achieved using the Sinkhorn algorithm， followed by refinement with the Iterative Closest Point （ICP） algorithm to enhance registration accuracy. Experimental results on the CODD and KITTI datasets， compared with various existing low-overlap point cloud registration methods， demonstrate that the network with ICP refinement performs superiorly. Specifically， on the CODD dataset， it reduces the Relative Translation Error （RTE） and Relative Rotation Error （RRE） by 53.29% and 42.72%， respectively， compared to the state-of-the-art method CoFiI2P （Coarse-to-Fine correspondences for Image-to-Point cloud registration）， while improving the Registration Recall （RR） by 0.2 percentage points. The results indicate that the proposed network effectively extracts descriptive information from keypoint features and significantly improves point cloud registration accuracy in low-overlap scenarios.

Key words: low-overlap scenario, point cloud registration, position encoding, overlap mask, self-attention, cross-attention

摘要：

针对低重叠场景下点云配准方法存在的关键点特征描述信息不足和重叠点云区域较少，进而导致点云的误匹配率高以及配准精度低的问题，设计一种融合位置编码和重叠掩模的低重叠点云配准网络，以降低点云的误匹配率，并提高配准的精度。首先，采用PointNet逐点特征编码器提取点云关键点，并融合关键点的特征信息、坐标信息和位置编码，生成更具判别力的关键点特征；其次，将融合后的特征输入自注意力和交叉注意力模块，以增强点云特征的描述能力，加强点云的上下文信息交互，从而解决关键点描述信息不足的问题；再次，在注意力模块之后引入重叠掩模模块，通过学习重叠掩模去除非重叠区域的关键点，以降低误匹配率；最后，结合Sinkhorn算法进行最优匹配，并采用迭代最近邻点（ICP）算法进行细化，提高点云配准精度。在CODD数据集和KITTI数据集上与多种现有的低重叠点云配准方法进行对比的实验结果表明，经过ICP细化后的网络性能更优，特别是在CODD数据集上，它比当前先进的低重叠点云配准方法CoFiI2P （Coarse-to-Fine correspondences for Image-to-Point cloud registration）的相对平移误差（RTE）和相对旋转误差（RRE）分别降低了53.29%和42.72%，配准召回率（RR）提升了0.2个百分点。可见，该网络能充分提取关键点特征的描述信息，并有效提升低重叠场景下的点云配准精度。

关键词: 低重叠场景, 点云配准, 位置编码, 重叠掩模, 自注意力, 交叉注意力

CLC Number:

TP391.41

Xiaowei LA, Lihua HU, Jianhua HU, Xiaoling YAO, Xinbo WANG. Low-overlap point cloud registration network integrating position encoding and overlap masks[J]. Journal of Computer Applications, 2026, 46(2): 536-545.

喇孝伟, 胡立华, 胡建华, 姚晓玲, 王欣波. 融合位置编码和重叠掩模的低重叠点云配准网络[J]. 《计算机应用》唯一官方网站, 2026, 46(2): 536-545.

Figures/Tables 12

Fig. 1 Structure of baseline network

Fig. 2 Framework for low-overlap point cloud registration network

Fig. 3 Structure of core module of point cloud registration network

Fig. 4 Structure of overlap mask module

Tab. 1 Information of KITTI and CODD datasets

数据集	子集	序列数	样本数
KITTI	训练集	5	1 358
	验证集	2	180
	测试集	2	255
CODD	训练集	78	6 129
	验证集	14	1 339
	测试集	15	1 315

Tab. 2 Impact of distance threshold on registration recall on KITTI and CODD datasets

$δ$ 取值/m	配准召回率/%
$δ$ 取值/m	KITTI	CODD
0.2	97.3	94.5
0.3	98.3	95.0
0.4	97.0	94.1
0.5	96.4	93.4

Tab. 2 Impact of distance threshold on registration recall on KITTI and CODD datasets

$δ$ 取值/m	配准召回率/%
$δ$ 取值/m	KITTI	CODD
0.2	97.3	94.5
0.3	98.3	95.0
0.4	97.0	94.1
0.5	96.4	93.4

Tab. 3 Impact of voxel on registration recall on KITTI and CODD datasets

$v s$ 取值/m	配准召回率/%
$v s$ 取值/m	KITTI	CODD
0.1	96.6	94.0
0.2	97.4	94.7
0.3	98.3	95.0
0.4	97.0	94.1

Tab. 3 Impact of voxel on registration recall on KITTI and CODD datasets

$v s$ 取值/m	配准召回率/%
$v s$ 取值/m	KITTI	CODD
0.1	96.6	94.0
0.2	97.4	94.7
0.3	98.3	95.0
0.4	97.0	94.1

Tab. 4 Experimental results on KITTI dataset

模型	RTE/m （↓）	RRE/（°）（↓）	RR/%（↑）
ICP	6.124	5.890	8.6
Symmetric ICP	5.340	5.260	12.4
FPFH	0.265	0.890	97.8
TEASER	0.190	0.750	98.6
FCGF	0.237	0.300	97.5
DGR	0.156	0.370	98.2
3DFeatNet	0.259	0.250	96.0
SpinNet	0.099	0.470	99.1
Predator	0.068	0.270	99.8
CoFiNet	0.082	0.410	99.8
HRegNet	0.120	0.290	99.7
RegFormer	0.086	0.232	99.8
CoFiI2P	0.073	0.260	99.7
基线网络	0.261	0.738	97.0
基线网络+ICP	0.092	0.254	98.5
本文网络	0.233	0.614	98.3
本文网络+ICP	0.066	0.178	99.2

Tab. 5 Experimental results on CODD dataset

模型	RTE/m （↓）	RRE/（°）（↓）	RR/%（↑）
ICP	16.136	74.842	7.5
Symmetric ICP	15.106	73.551	12.8
FPFH	13.802	69.593	31.3
TEASER	12.908	69.208	38.0
FCGF	1.700	0.180	91.0
DGR	0.390	1.520	94.0
RegFormer	0.135	0.189	97.4
CoFiI2P	0.167	0.213	97.6
基线网络	0.313	0.415	92.7
基线网络+ICP	0.120	0.130	97.0
本文网络	0.261	0.357	95.0
本文网络+ICP	0.078	0.122	97.8

Tab. 6 Ablation experimental results on KITTI and CODD datasets

模型	KITTI			CODD
模型	RTE/m （↓）	RRE/（°）	RR/%	RTE/m	RRE/（°）	RR/%
基线模型	0.261	0.738	97.0	0.313	0.415	91.7
基线模型+位置编码	0.239	0.645	97.5	0.290	0.356	94.7
基线模型+重叠掩模	0.245	0.656	97.4	0.283	0.371	94.5
基线模型+位置编码+重叠掩模	0.235	0.630	97.7	0.290	0.382	95.0
基线模型+位置编码+重叠掩模+Sinkhorn	0.233	0.614	98.3	0.261	0.357	95.0

Fig. 5 Visualization results on KITTI dataset

Fig. 6 Visualization results on CODD dataset

References 45

[1]	AZUMA R T. A survey of augmented reality ［J］. Presence： Teleoperators and Virtual Environments， 1997， 6（4）： 355-385.
[2]	CARMIGNIANI J， FURHT B， ANISETTI M， et al. Augmented reality technologies， systems and applications ［J］. Multimedia Tools and Applications， 2011， 51（1）： 341-377.
[3]	徐帅.基于点云配准的单目视觉多人增强现实机制的研究［D］.北京：北京邮电大学， 2020： 22-45.
	XU S. Research on monocular vision multi-user augmented reality based on point cloud registration ［D］. Beijing： Beijing University of Posts and Telecommunications， 2020： 22-45.
[4]	IZADI S， KIM D， HILLIGES O， et al. KinectFusion： real-time 3D reconstruction and interaction using a moving depth camera ［C］// Proceedings of the 24th Annual ACM Symposium on User Interface Software and Technology. New York： ACM， 2011： 559-568.
[5]	NIEβNER M， ZOLLHÖFER M， IZADI S， et al. Real-time 3D reconstruction at scale using voxel hashing ［J］. ACM Transactions on Graphics， 2013， 32（6）： No.169.
[6]	刘建斌，马燕，黄慧.基于快速聚合和预配准的三维人脸重建［J］.计算机与数字工程， 2024， 52（5）： 1477-1481.
	LIU J B， MA Y， HUANG H. 3D face reconstruction based on fast aggregation and pre-registration ［J］. Computer and Digital Engineering， 2024， 52（5）： 1477-1481.
[7]	GEIGER A， LENZ P， URTASUN R. Are we ready for autonomous driving？ the KITTI vision benchmark suite ［C］// Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2012： 3354-3361.
[8]	YURTSEVER E， LAMBERT J， CARBALLO A， et al. A survey of autonomous driving： common practices and emerging technologies ［J］. IEEE Access， 2020， 8： 58443-58469.
[9]	柳斌.面向自动驾驶的MEMS激光雷达目标点云配准方法研究［D］.吉林：吉林化工学院， 2024： 18-31.
	LIU B. Research on MEMS LiDAR target point cloud registration method for autonomous driving ［D］. Jilin： Jilin Institute of Chemical Technology， 2024： 18-31.
[10]	BESL P J， McKAY N D. Method for registration of 3-D shapes ［C］// Proceedings of the SPIE 1611， Sensor Fusion IV： Control Paradigms and Data Structures. Bellingham， WA： SPIE， 1992： 586-606.
[11]	RUSU R B， BLODOW N， BEETZ M. Fast Point Feature Histograms （FPFH） for 3D registration ［C］// Proceedings of the 2009 IEEE International Conference on Robotics and Automation. Piscataway： IEEE， 2009： 3212-3217.
[12]	YANG J， LI H， JIA Y. Go-ICP： solving 3D registration efficiently and globally optimally ［C］// Proceedings of the 2013 IEEE International Conference on Computer Vision. Piscataway： IEEE， 2013： 1457-1464.
[13]	RUSINKIEWICZ S. A symmetric objective function for ICP ［J］. ACM Transactions on Graphics， 2019， 38（4）： No.85.
[14]	LIU H， ZHANG Y， LEI L， et al. Hierarchical optimization of 3D point cloud registration ［J］. Sensors， 2020， 20（23）： No.6999.
[15]	YANG H， SHI J， CARLONE L. TEASER： fast and certifiable point cloud registration ［J］. IEEE Transactions on Robotics， 2021， 37（2）： 314-333.
[16]	朱乾荣，柏艳红，王银，等.基于3D形状上下文特征改进的ICP点云配准算法［J］.海洋测绘， 2023， 43（2）： 69-73.
	ZHU Q R， BAI Y H， WANG Y， et al. Improved ICP point cloud registration algorithm based on 3D shape context features ［J］. Hydrographic Surveying and Charting， 2023， 43（2）： 69-73.
[17]	唐佳林，林寿南，周壮，等.基于特征变换结合KD树改进ICP的快速点云配准方法［J］.计算机科学， 2023， 50（11A）： No.230100028.
	TANG J L， LIN S N， ZHOU Z， et al. Improved ICP fast point cloud registration method based on feature transformation combined with KD tree ［J］. Computer Science， 2023， 50（11A）： No.230100028.
[18]	AOKI Y， GOFORTH H， SRIVATSAN R A， et al. PointNetLK： robust and efficient point cloud registration using PointNet ［C］// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2019： 7156-7165.
[19]	QI C R， SU H， MO K， et al. PointNet： deep learning on point sets for 3D classification and segmentation ［C］// Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2017： 77-85.
[20]	LUCAS B D， KANADE T. An iterative image registration technique with an application to stereo vision ［C］// Proceedings of the 7th International Joint Conference on Artificial Intelligence — Volume 2. San Francisco： Morgan Kaufmann Publishers Inc.， 1981： 674-679.
[21]	SARODE V， LI X， GOFORTH H， et al. PCRNet： point cloud registration network using PointNet encoding ［EB/OL］. ［2024-08-07］. .
[22]	WANG Y， SOLOMON J M. Deep closest point： learning representations for point cloud registration ［C］// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2019： 3522-3531.
[23]	WANG Y， SUN Y， LIU Z， et al. Dynamic graph CNN for learning on point clouds ［J］. ACM Transactions on Graphics， 2019， 38（5）： No.146.
[24]	ARUN K S， HUANG T S， BLOSTEIN S D. Least-squares fitting of two 3-D point sets ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 1987， PAMI-9（5）： 698-700.
[25]	WANG Y， SOLOMON J. PRNet： self-supervised learning for partial-to-partial registration ［C］// Proceedings of the 33rd International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2019： 8814-8826.
[26]	YEW Z J， LEE G H. RPM-Net： robust point matching using learned features ［C］// Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2020： 11821-11830.
[27]	SINKHORN R. A relationship between arbitrary positive matrices and doubly stochastic matrices ［J］. The Annals of Mathematical Statistics， 1964， 35（2）： 876-879.
[28]	YUAN Y， BORRMANN D， HOU J， et al. Self-supervised point set local descriptors for point cloud registration ［J］. Sensors， 2021， 21（2）： No.486.
[29]	YEW Z J， LEE G H. 3DFeat-Net： weakly supervised local 3D features for point cloud registration ［C］// Proceedings of the 2018 European Conference on Computer Vision， LNCS 11219. Cham： Springer， 2018： 630-646.
[30]	AO S， HU Q， YANG B， et al. SpinNet： learning a general surface descriptor for 3D point cloud registration ［C］// Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2021： 11748-11757.
[31]	YI R， LI J， LUO L， et al. DOPNet： achieving accurate and efficient point cloud registration based on deep learning and multi-level features ［J］. Sensors， 2022， 22（21）： No.8217.
[32]	KANG S， LIAO Y， LI J， et al. CoFiI2P： coarse-to-fine correspondences for image-to-point cloud registration ［J］. IEEE Robotics and Automation Letters， 2024， 9（11）： 10264-10271.
[33]	LU F， CHEN G， LIU Y， et al. HRegNet： a hierarchical network for efficient and accurate outdoor LiDAR point cloud registration ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2023， 45（10）： 11884-11897.
[34]	徐嘉伯.用于低重叠点云配准任务的强泛化全卷积神经网络［D］.南昌：南昌大学， 2023： 9-29.
	XU J B. Strong generalization fully convolutional neural network for low-overlap point cloud registration ［D］. Nanchang： Nanchang University， 2023： 9-29.
[35]	LIU J， WANG G， LIU Z， et al. RegFormer： an efficient projection-aware Transformer network for large-scale point cloud registration ［C］// Proceedings of the 2023 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2023： 8417-8426.
[36]	DENG H， BIRDAL T， ILIC S. PPFNet： global context aware local features for robust 3D point matching ［C］// Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2018： 195-205.
[37]	FISCHER M A， BOLLES R C. Random sample consensus： a paradigm for model fitting with applications to image analysis and automated cartography ［J］. Communications of the ACM， 1981， 24（6）： 381-395.
[38]	CHOY C， PARK J， KOLTUN V. Fully convolutional geometric features ［C］// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2019： 8957-8965.
[39]	CHOY C， DONG W， KOLTUN V. Deep global registration ［C］// Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2020： 2511-2520.
[40]	HUANG S， GOJCIC Z， USVYATSOV M， et al. PREDATOR： registration of 3D point clouds with low overlap ［C］// Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2021： 4265-4274.
[41]	YU H， LI F， SALEH M， et al. CoFiNet： reliable coarse-to-fine correspondences for robust point cloud registration ［C］// Proceedings of the 35th International Conference on Neural Information Processing Systems. Red Hook： Curran Associates Inc.， 2021： 23872-23884.
[42]	ARNOLD E， MOZAFFARI S， DIANATI M. Fast and robust registration of partially overlapping point clouds ［J］. IEEE Robotics and Automation Letters， 2022， 7（2）： 1502-1509.
[43]	QIN Z， YU H， WANG C， et al. Geometric Transformer for fast and robust point cloud registration ［C］// Proceedings of the 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2022： 11133-11142.
[44]	刘旭珩，柏正尧，许祝，等.结合注意力机制的多重引导点云配准网络［J］.计算机科学， 2024， 51（2）： 142-150.
	LIU X H， BAI Z Y， XU Z， et al. Multi-guided point cloud registration network combined with attention mechanism ［J］. Computer Science， 2024， 51（2）： 142-150.
[45]	杜佳锦，柏正尧，刘旭珩，等.融合几何注意力和多尺度特征点云配准网络［J］.计算机工程与应用， 2024， 60（12）： 234-244.
	DU J J， BAI Z Y， LIU X H， et al. Fusion of geometric attention and multi-scale feature network for point cloud registration ［J］. Computer Engineering and Applications， 2024， 60（12）： 234-244.

Low-overlap point cloud registration network integrating position encoding and overlap masks

融合位置编码和重叠掩模的低重叠点云配准网络

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 12

References 45

Related Articles 15

Recommended Articles

Metrics

[1]	Hu LUO, Mingshu ZHANG. Rumor detection method based on cross-modal attention mechanism and contrastive learning [J]. Journal of Computer Applications, 2026, 46(2): 361-367.
[2]	Yanan LI, Mengyang GUO, Guojun DENG, Yunfeng CHEN, Jianji REN, Yongliang YUAN. Method for life prediction of parallel branching engine based on multi-modal fusion features [J]. Journal of Computer Applications, 2026, 46(1): 305-313.
[3]	Fan HE, Li LI, Zhongxu YUAN, Xiu YANG, Dongxuan HAN. Knowledge tracking model based on concept association memory network with graph attention [J]. Journal of Computer Applications, 2026, 46(1): 43-51.
[4]	Xiang WANG, Zhixiang CHEN, Guojun MAO. Multivariate time series prediction method combining local and global correlation [J]. Journal of Computer Applications, 2025, 45(9): 2806-2816.
[5]	Yilin DENG, Fajiang YU. Pseudo random number generator based on LSTM and separable self-attention mechanism [J]. Journal of Computer Applications, 2025, 45(9): 2893-2901.
[6]	Chen LIANG, Yisen WANG, Qiang WEI, Jiang DU. Source code vulnerability detection method based on Transformer-GCN [J]. Journal of Computer Applications, 2025, 45(7): 2296-2303.
[7]	Yihan WANG, Chong LU, Zhongyuan CHEN. Multimodal sentiment analysis model with cross-modal text information enhancement [J]. Journal of Computer Applications, 2025, 45(7): 2237-2244.
[8]	Xiaoqiang ZHAO, Yongyong LIU, Yongyong HUI, Kai LIU. Batch process quality prediction model using improved time-domain convolutional network with multi-head self-attention mechanism [J]. Journal of Computer Applications, 2025, 45(7): 2245-2252.
[9]	Mingfeng YU, Yongbin QIN, Ruizhang HUANG, Yanping CHEN, Chuan LIN. Multi-label text classification method based on contrastive learning enhanced dual-attention mechanism [J]. Journal of Computer Applications, 2025, 45(6): 1732-1740.
[10]	Hui LI, Bingzhi JIA, Chenxi WANG, Ziyu DONG, Jilong LI, Zhaoman ZHONG, Yanyan CHEN. Generative adversarial network underwater image enhancement model based on Swin Transformer [J]. Journal of Computer Applications, 2025, 45(5): 1439-1446.
[11]	Kunyuan JIANG, Xiaoxia LI, Li WANG, Yaodan CAO, Xiaoqiang ZHANG, Nan DING, Yingyue ZHOU. Boundary-cross supervised semantic segmentation network with decoupled residual self-attention [J]. Journal of Computer Applications, 2025, 45(4): 1120-1129.
[12]	Weigang LI, Wenjie CAO, Jinling LI. Multi-stage point cloud completion network based on adaptive neighborhood feature fusion [J]. Journal of Computer Applications, 2025, 45(10): 3294-3301.
[13]	Hu WANG, Xiaofeng WANG, Ke LI, Yunjie MA. Federated class-incremental learning method of label semantic embedding with multi-head self-attention [J]. Journal of Computer Applications, 2025, 45(10): 3083-3090.
[14]	Jietao LIANG, Bing LUO, Lanhui FU, Qingling CHANG, Nannan LI, Ningbo YI, Qi FENG, Xin HE, Fuqin DENG. Point cloud registration method based on coordinate geometric sampling [J]. Journal of Computer Applications, 2025, 45(1): 214-222.
[15]	Pengcheng SONG, Lijun GUO, Rong ZHANG. Weakly supervised video anomaly detection with local-global temporal dependency [J]. Journal of Computer Applications, 2025, 45(1): 240-246.