Mofei Song
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Mofei Song has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Zhengxing Sun
State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, Jiangsu, China
7 shared publications Feiqian Zhang
Nanjing University
3 shared publications- Lv Tang
School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing, China
3 shared publications Xufeng Lang
Nanjing University
3 shared publicationsBo Li
YouTu Lab, Tencent, Shanghai, China
2 shared publicationsQian Li
College of Meteorology and Oceanography, National University of Defense Technology, Changsha, China
2 shared publications
Latest Publications
- 5 May 2026
- Published by Elsevier in Pattern Recognition
- Vol. 179
3D semi-supervised classification based on multi-modal data augmentation and self-adaptive thresholding
- 5 May 2026
- Published by Elsevier in Pattern Recognition
- Vol. 179
3D semi-supervised classification based on multi-modal data augmentation and self-adaptive thresholding
- 22 March 2026
- Published by Springer Nature in The Visual Computer
- Vol. 42 (5) , 218
UniGS: unified geometry reconstruction for specular and semi-transparent surfaces via 3D Gaussian splatting
- 22 March 2026
- Published by Springer Nature in The Visual Computer
- Vol. 42 (5) , 218
UniGS: unified geometry reconstruction for specular and semi-transparent surfaces via 3D Gaussian splatting
- 12 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 37 (11) , 6694-6707
Similarity and Dissimilarity Guided Co-Association Matrix Construction for Ensemble Clustering
Ensemble clustering aggregates multiple weak clusterings to achieve a more accurate and robust consensus result. The Co-Association matrix (CA matrix) based method is the mainstream ensemble clustering approach that constructs the similarity relationships between sample pairs according the weak clustering partitions to generate the final clustering result. However, the existing methods neglect that the quality of cluster is related to its size, i.e., a cluster with smaller size tends to higher accuracy. Moreover, they also do not consider the valuable dissimilarity information in the base clusterings which can reflect the varying importance of sample pairs that are completely disconnected. To this end, we propose the Similarity and Dissimilarity Guided Co-association matrix (SDGCA) to achieve ensemble clustering. First, we introduce normalized ensemble entropy to estimate the quality of each cluster, and construct a similarity matrix based on this estimation. Then, we employ the random walk to explore high-order proximity of base clusterings to construct a dissimilarity matrix. Finally, the adversarial relationship between the similarity matrix and the dissimilarity matrix is utilized to construct a promoted CA matrix for ensemble clustering. We compared our method with 13 state-of-the-art methods across 12 datasets, and the results demonstrated the superior clustering ability and robustness of the proposed approach. The code is available at https://github.com/xuz2019/SDGCA.
- 12 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 37 (11) , 6694-6707
Similarity and Dissimilarity Guided Co-Association Matrix Construction for Ensemble Clustering
Ensemble clustering aggregates multiple weak clusterings to achieve a more accurate and robust consensus result. The Co-Association matrix (CA matrix) based method is the mainstream ensemble clustering approach that constructs the similarity relationships between sample pairs according the weak clustering partitions to generate the final clustering result. However, the existing methods neglect that the quality of cluster is related to its size, i.e., a cluster with smaller size tends to higher accuracy. Moreover, they also do not consider the valuable dissimilarity information in the base clusterings which can reflect the varying importance of sample pairs that are completely disconnected. To this end, we propose the Similarity and Dissimilarity Guided Co-association matrix (SDGCA) to achieve ensemble clustering. First, we introduce normalized ensemble entropy to estimate the quality of each cluster, and construct a similarity matrix based on this estimation. Then, we employ the random walk to explore high-order proximity of base clusterings to construct a dissimilarity matrix. Finally, the adversarial relationship between the similarity matrix and the dissimilarity matrix is utilized to construct a promoted CA matrix for ensemble clustering. We compared our method with 13 state-of-the-art methods across 12 datasets, and the results demonstrated the superior clustering ability and robustness of the proposed approach. The code is available at https://github.com/xuz2019/SDGCA.
- 3 July 2025
- Published by Elsevier in Pattern Recognition Letters
- Vol. 196, 303-311
Cross-domain 3D model classification via pseudo-labeling noise correction
- 3 July 2025
- Published by Elsevier in Pattern Recognition Letters
- Vol. 196, 303-311
Cross-domain 3D model classification via pseudo-labeling noise correction
- Open Access
Similarity and Dissimilarity Guided Co-association Matrix Construction for Ensemble Clustering
Ensemble clustering aggregates multiple weak clusterings to achieve a more accurate and robust consensus result. The Co-Association matrix (CA matrix) based method is the mainstream ensemble clustering approach that constructs the similarity relationships between sample pairs according the weak clustering partitions to generate the final clustering result. However, the existing methods neglect that the quality of cluster is related to its size, i.e., a cluster with smaller size tends to higher accuracy. Moreover, they also do not consider the valuable dissimilarity information in the base clusterings which can reflect the varying importance of sample pairs that are completely disconnected. To this end, we propose the Similarity and Dissimilarity Guided Co-association matrix (SDGCA) to achieve ensemble clustering. First, we introduce normalized ensemble entropy to estimate the quality of each cluster, and construct a similarity matrix based on this estimation. Then, we employ the random walk to explore high-order proximity of base clusterings to construct a dissimilarity matrix. Finally, the adversarial relationship between the similarity matrix and the dissimilarity matrix is utilized to construct a promoted CA matrix for ensemble clustering. We compared our method with 13 state-of-the-art methods across 12 datasets, and the results demonstrated the superiority clustering ability and robustness of the proposed approach. The code is available at https://github.com/xuz2019/SDGCA.
Open AccessSimilarity and Dissimilarity Guided Co-association Matrix Construction for Ensemble Clustering
Ensemble clustering aggregates multiple weak clusterings to achieve a more accurate and robust consensus result. The Co-Association matrix (CA matrix) based method is the mainstream ensemble clustering approach that constructs the similarity relationships between sample pairs according the weak clustering partitions to generate the final clustering result. However, the existing methods neglect that the quality of cluster is related to its size, i.e., a cluster with smaller size tends to higher accuracy. Moreover, they also do not consider the valuable dissimilarity information in the base clusterings which can reflect the varying importance of sample pairs that are completely disconnected. To this end, we propose the Similarity and Dissimilarity Guided Co-association matrix (SDGCA) to achieve ensemble clustering. First, we introduce normalized ensemble entropy to estimate the quality of each cluster, and construct a similarity matrix based on this estimation. Then, we employ the random walk to explore high-order proximity of base clusterings to construct a dissimilarity matrix. Finally, the adversarial relationship between the similarity matrix and the dissimilarity matrix is utilized to construct a promoted CA matrix for ensemble clustering. We compared our method with 13 state-of-the-art methods across 12 datasets, and the results demonstrated the superiority clustering ability and robustness of the proposed approach. The code is available at https://github.com/xuz2019/SDGCA.
- Open Access
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 8355-8364
UniL: Point Cloud Novelty Detection through Multimodal Pre-training
3D novelty detection plays a crucial role in various real-world applications, especially in safety-critical fields such as autonomous driving and intelligent surveillance systems. However, existing 3D novelty detection methods are constrained by the scarcity of 3D data, which may impede the model's ability to learn adequate representations, thereby impacting detection accuracy. To address this challenge, we propose a Unified Learning Framework (UniL) for facilitating novelty detection. During the pretraining phase, UniL assists the point cloud encoder in learning information from other modalities, aligning visual, textual, and 3D features within the same feature space. Additionally, we introduce a novel Multimodal Supervised Contrastive Loss (MSC Loss) to improve the model's ability to cluster samples from the same category in feature space by leveraging label information during pretraining. Furthermore, we propose a straightforward yet powerful scoring method, Depth Map Error (DME), which assesses the discrepancy between projected depth maps before and after point cloud reconstruction during novelty detection. Extensive experiments conducted on 3DOS have demonstrated the effectiveness of our approach, significantly enhancing the performance of the unsupervised VAE method in 3D novelty detection. Codes are avaliable at https://github.com/EugeneWon9/UniL.
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 8355-8364
Open AccessUniL: Point Cloud Novelty Detection through Multimodal Pre-training
3D novelty detection plays a crucial role in various real-world applications, especially in safety-critical fields such as autonomous driving and intelligent surveillance systems. However, existing 3D novelty detection methods are constrained by the scarcity of 3D data, which may impede the model's ability to learn adequate representations, thereby impacting detection accuracy. To address this challenge, we propose a Unified Learning Framework (UniL) for facilitating novelty detection. During the pretraining phase, UniL assists the point cloud encoder in learning information from other modalities, aligning visual, textual, and 3D features within the same feature space. Additionally, we introduce a novel Multimodal Supervised Contrastive Loss (MSC Loss) to improve the model's ability to cluster samples from the same category in feature space by leveraging label information during pretraining. Furthermore, we propose a straightforward yet powerful scoring method, Depth Map Error (DME), which assesses the discrepancy between projected depth maps before and after point cloud reconstruction during novelty detection. Extensive experiments conducted on 3DOS have demonstrated the effectiveness of our approach, significantly enhancing the performance of the unsupervised VAE method in 3D novelty detection. Codes are avaliable at https://github.com/EugeneWon9/UniL.
- Open Access
- 24 March 2024
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 38 (7) , 6773-6781
MM-Point: Multi-View Information-Enhanced Multi-Modal Self-Supervised 3D Point Cloud Understanding
In perception, multiple sensory information is integrated to map visual information from 2D views onto 3D objects, which is beneficial for understanding in 3D environments. But in terms of a single 2D view rendered from different angles, only limited partial information can be provided. The richness and value of Multi-view 2D information can provide superior self-supervised signals for 3D objects. In this paper, we propose a novel self-supervised point cloud representation learning method, MM-Point, which is driven by intra-modal and inter-modal similarity objectives. The core of MM-Point lies in the Multi-modal interaction and transmission between 3D objects and multiple 2D views at the same time. In order to more effectively simultaneously perform the consistent cross-modal objective of 2D multi-view information based on contrastive learning, we further propose Multi-MLP and Multi-level Augmentation strategies. Through carefully designed transformation strategies, we further learn Multi-level invariance in 2D Multi-views. MM-Point demonstrates state-of-the-art (SOTA) performance in various downstream tasks. For instance, it achieves a peak accuracy of 92.4% on the synthetic dataset ModelNet40, and a top accuracy of 87.8% on the real-world dataset ScanObjectNN, comparable to fully supervised methods. Additionally, we demonstrate its effectiveness in tasks such as few-shot classification, 3D part segmentation and 3D semantic segmentation.
- 24 March 2024
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 38 (7) , 6773-6781
Open AccessMM-Point: Multi-View Information-Enhanced Multi-Modal Self-Supervised 3D Point Cloud Understanding
In perception, multiple sensory information is integrated to map visual information from 2D views onto 3D objects, which is beneficial for understanding in 3D environments. But in terms of a single 2D view rendered from different angles, only limited partial information can be provided. The richness and value of Multi-view 2D information can provide superior self-supervised signals for 3D objects. In this paper, we propose a novel self-supervised point cloud representation learning method, MM-Point, which is driven by intra-modal and inter-modal similarity objectives. The core of MM-Point lies in the Multi-modal interaction and transmission between 3D objects and multiple 2D views at the same time. In order to more effectively simultaneously perform the consistent cross-modal objective of 2D multi-view information based on contrastive learning, we further propose Multi-MLP and Multi-level Augmentation strategies. Through carefully designed transformation strategies, we further learn Multi-level invariance in 2D Multi-views. MM-Point demonstrates state-of-the-art (SOTA) performance in various downstream tasks. For instance, it achieves a peak accuracy of 92.4% on the synthetic dataset ModelNet40, and a top accuracy of 87.8% on the real-world dataset ScanObjectNN, comparable to fully supervised methods. Additionally, we demonstrate its effectiveness in tasks such as few-shot classification, 3D part segmentation and 3D semantic segmentation.
- Open Access
- 1 January 2023
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 75 (1) , 197-217
RankXGB-Based Enterprise Credit Scoring by Electricity Consumption in Edge Computing Environment
With the rapid development of the internet of things (IoT), electricity consumption data can be captured and recorded in the IoT cloud center. This provides a credible data source for enterprise credit scoring, which is one of the most vital elements during the financial decision-making process. Accordingly, this paper proposes to use deep learning to train an enterprise credit scoring model by inputting the electricity consumption data. Instead of predicting the credit rating, our method can generate an absolute credit score by a novel deep ranking model–ranking extreme gradient boosting net (rankXGB). To boost the performance, the rankXGB model combines several weak ranking models into a strong model. Due to the high computational cost and the vast amounts of data, we design an edge computing framework to reduce the latency of enterprise credit evaluation. Specially, we design a two-stage deep learning task architecture, including a cloud-based weak credit ranking and an edge-based credit score calculation. In the first stage, we send the electricity consumption data of the evaluated enterprise to the computing cloud server, where multiple weak-ranking networks are executed in parallel to produce multiple weak-ranking results. In the second stage, the edge device fuses multiple ranking results generated in the cloud server to produce a more reliable ranking result, which is used to calculate an absolute credit score by score normalization. The experiments demonstrate that our method can achieve accurate enterprise credit evaluation quickly.
- 1 January 2023
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 75 (1) , 197-217
Open AccessRankXGB-Based Enterprise Credit Scoring by Electricity Consumption in Edge Computing Environment
With the rapid development of the internet of things (IoT), electricity consumption data can be captured and recorded in the IoT cloud center. This provides a credible data source for enterprise credit scoring, which is one of the most vital elements during the financial decision-making process. Accordingly, this paper proposes to use deep learning to train an enterprise credit scoring model by inputting the electricity consumption data. Instead of predicting the credit rating, our method can generate an absolute credit score by a novel deep ranking model–ranking extreme gradient boosting net (rankXGB). To boost the performance, the rankXGB model combines several weak ranking models into a strong model. Due to the high computational cost and the vast amounts of data, we design an edge computing framework to reduce the latency of enterprise credit evaluation. Specially, we design a two-stage deep learning task architecture, including a cloud-based weak credit ranking and an edge-based credit score calculation. In the first stage, we send the electricity consumption data of the evaluated enterprise to the computing cloud server, where multiple weak-ranking networks are executed in parallel to produce multiple weak-ranking results. In the second stage, the edge device fuses multiple ranking results generated in the cloud server to produce a more reliable ranking result, which is used to calculate an absolute credit score by score normalization. The experiments demonstrate that our method can achieve accurate enterprise credit evaluation quickly.
- Open Access
- 10 December 2022
- Published by Springer Nature in Journal of Cloud Computing
- Vol. 11 (1) , 1-16
Efficient 3D object recognition in mobile edge environment
- 10 December 2022
- Published by Springer Nature in Journal of Cloud Computing
- Vol. 11 (1) , 1-16
Open AccessEfficient 3D object recognition in mobile edge environment
- 13 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 31 (10577149) , 6532-6547
Toward Stable Co-Saliency Detection and Object Co-Segmentation
In this paper, we present a novel model for simultaneous stable co-saliency detection (CoSOD) and object co-segmentation (CoSEG). To detect co-saliency (segmentation) accurately, the core problem is to well model inter-image relations between an image group. Some methods design sophisticated modules, such as recurrent neural network (RNN), to address this problem. However, order-sensitive problem is the major drawback of RNN, which heavily affects the stability of proposed CoSOD (CoSEG) model. In this paper, inspired by RNN-based model, we first propose a multi-path stable recurrent unit (MSRU), containing dummy orders mechanisms (DOM) and recurrent unit (RU). Our proposed MSRU not only helps CoSOD (CoSEG) model captures robust inter-image relations, but also reduces order-sensitivity, resulting in a more stable inference and training process. Moreover, we design a cross-order contrastive loss (COCL) that can further address order-sensitive problem by pulling close the feature embedding generated from different input orders. We validate our model on five widely used CoSOD datasets (CoCA, CoSOD3k, Cosal2015, iCoseg and MSRC), and three widely used datasets (Internet, iCoseg and PASCAL-VOC) for object co-segmentation, the performance demonstrates the superiority of the proposed approach as compared to the state-of-the-art (SOTA) methods.
- 13 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 31 (10577149) , 6532-6547
Toward Stable Co-Saliency Detection and Object Co-Segmentation
In this paper, we present a novel model for simultaneous stable co-saliency detection (CoSOD) and object co-segmentation (CoSEG). To detect co-saliency (segmentation) accurately, the core problem is to well model inter-image relations between an image group. Some methods design sophisticated modules, such as recurrent neural network (RNN), to address this problem. However, order-sensitive problem is the major drawback of RNN, which heavily affects the stability of proposed CoSOD (CoSEG) model. In this paper, inspired by RNN-based model, we first propose a multi-path stable recurrent unit (MSRU), containing dummy orders mechanisms (DOM) and recurrent unit (RU). Our proposed MSRU not only helps CoSOD (CoSEG) model captures robust inter-image relations, but also reduces order-sensitivity, resulting in a more stable inference and training process. Moreover, we design a cross-order contrastive loss (COCL) that can further address order-sensitive problem by pulling close the feature embedding generated from different input orders. We validate our model on five widely used CoSOD datasets (CoCA, CoSOD3k, Cosal2015, iCoseg and MSRC), and three widely used datasets (Internet, iCoseg and PASCAL-VOC) for object co-segmentation, the performance demonstrates the superiority of the proposed approach as compared to the state-of-the-art (SOTA) methods.