Zengfu Wang

EAAI Journal 2026 Journal Article

KalmanTNet - hybrid-driven Kalman filter for trajectory filtering

• Wenyi Zhang
• Zengfu Wang
• Kuilong Yang
• Hua Lan

The problem of trajectory filtering has garnered significant attention across various domains. However, existing methods struggle with complex situations and perform poorly under model mismatch conditions. To address these limitations, this paper proposes KalmanTNet, a novel data-model hybrid driven approach that leverages the robust sequence processing capabilities of the Transformer architecture and the concise structure of the Kalman filter. Experimental results show that KalmanTNet outperforms Kalman filter, extended Kalman filter, unscented Kalman filter and KalmanNet in complex environments, achieving superior filtering accuracy.

NeurIPS Conference 2025 Conference Paper

Unleashing Foundation Vision Models: Adaptive Transfer for Diverse Data-Limited Scientific Domains

• Qiankun Li
• Feng He
• Huabao Chen
• Xin Ning
• Kun Wang
• Zengfu Wang

In the big data era, the computer vision field benefits from large-scale datasets such as LAION-2B, LAION-400M, and ImageNet-21K, Kinetics, on which popular models like the ViT and ConvNeXt series have been pre-trained, acquiring substantial knowledge. However, numerous downstream tasks in specialized and data-limited scientific domains continue to pose significant challenges. In this paper, we propose a novel Cluster Attention Adapter (CLAdapter), which refines and adapts the rich representations learned from large-scale data to various data-limited downstream tasks. Specifically, CLAdapter introduces attention mechanisms and cluster centers to personalize the enhancement of transformed features through distribution correlation and transformation matrices. This enables models fine-tuned with CLAdapter to learn distinct representations tailored to different feature sets, facilitating the models' adaptation from rich pre-trained features to various downstream scenarios effectively. In addition, CLAdapter's unified interface design allows for seamless integration with multiple model architectures, including CNNs and Transformers, in both 2D and 3D contexts. Through extensive experiments on 10 datasets spanning domains such as generic, multimedia, biological, medical, industrial, agricultural, environmental, geographical, materials science, out-of-distribution (OOD), and 3D analysis, CLAdapter achieves state-of-the-art performance across diverse data-limited scientific domains, demonstrating its effectiveness in unleashing the potential of foundation vision models via adaptive transfer. Code is available at https: //github. com/qklee-lz/CLAdapter.

EAAI Journal 2024 Journal Article

A sea–land clutter classification framework for over-the-horizon radar based on weighted loss semi-supervised generative adversarial network

• Xiaoxuan Zhang
• Zengfu Wang
• Mingyue Ji
• Yang Li
• Quan Pan
• Kun Lu

Deep convolutional neural network has made great achievements in sea–land clutter classification for over-the-horizon radar (OTHR). The premise is that a large number of labeled training samples must be provided for a sea–land clutter classifier. In practical engineering applications, it is relatively easy to obtain label-free sea–land clutter samples. However, the labeling process is extremely cumbersome and requires expertise in the field of OTHR. To solve this problem, we propose an improved generative adversarial network, namely weighted loss semi-supervised generative adversarial network (WL-SSGAN). Specifically, we propose a joint feature matching loss by weighting the middle layer features of the discriminator of semi-supervised generative adversarial network (SSGAN). Furthermore, we propose the weighted loss of WL-SSGAN by linearly weighting the standard adversarial loss of SSGAN and the joint feature matching loss. The classification performance of WL-SSGAN is evaluated on sea–land clutter datasets. The experimental results show that WL-SSGAN can improve the performance of the fully supervised classifier with only a small number of labeled samples by utilizing a large number of unlabeled sea–land clutter samples. Further, the proposed weighted loss is superior to both the adversarial loss and the feature matching loss. Additionally, we compare WL-SSGAN with conventional semi-supervised classification methods and demonstrate that WL-SSGAN achieves the highest classification accuracy.

ICLR Conference 2023 Conference Paper

Optimizing Spca-based Continual Learning: A Theoretical Approach

• Chunchun Yang
• Malik Tiomoko
• Zengfu Wang

Catastrophic forgetting and the stability-plasticity dilemma are two major obstacles to continual learning. In this paper we first propose a theoretical analysis of a SPCA-based continual learning algorithm using high dimensional statistics. Second, we design OSCL (Optimized Spca-based Continual Learning) which builds on a flexible task optimization based on the theory. By optimizing a single task, catastrophic forgetting can be prevented theoretically. While optimizing multi-tasks, the trade-off between integrating knowledge from the new task and retaining previous knowledge of the old task can be achieved by assigning appropriate weights to corresponding tasks in compliance with the objectives. Experimental results confirm that the various theoretical conclusions are robust to a wide range of data distributions. Besides, several applications on synthetic and real data show that the proposed method while being computationally efficient, achieves comparable results with some state of the art.

AAAI Conference 2020 Conference Paper

Simple Pose: Rethinking and Improving a Bottom-up Approach for Multi-Person Pose Estimation

• Jia Li
• Wen Su
• Zengfu Wang

We rethink a well-known bottom-up approach for multiperson pose estimation and propose an improved one. The improved approach surpasses the baseline significantly thanks to (1) an intuitional yet more sensible representation, which we refer to as body parts to encode the connection information between keypoints, (2) an improved stacked hourglass network with attention mechanisms, (3) a novel focal L2 loss which is dedicated to “hard” keypoint and keypoint association (body part) mining, and (4) a robust greedy keypoint assignment algorithm for grouping the detected keypoints into individual poses. Our approach not only works straightforwardly but also outperforms the baseline by about 15% in average precision and is comparable to the state of the art on the MS-COCO test-dev dataset. The code and pre-trained models are publicly available on our project page1.

IJCAI Conference 2020 Conference Paper

Weakly Supervised Local-Global Relation Network for Facial Expression Recognition

• Haifeng Zhang
• Wen Su
• Jun Yu
• Zengfu Wang

To extract crucial local features and enhance the complementary relation between local and global features, this paper proposes a Weakly Supervised Local-Global Relation Network (WS-LGRN), which uses the attention mechanism to deal with part location and feature fusion problems. Firstly, the Attention Map Generator quickly finds the local regions-of-interest under the supervision of image-level labels. Secondly, bilinear attention pooling is employed to generate and refine local features. Thirdly, Relational Reasoning Unit is designed to model the relation among all features before making classification. The weighted fusion mechanism in the Relational Reasoning Unit makes the model benefit from the complementary advantages between different features. In addition, contrastive losses are introduced for local and global features to increase the inter-class dispersion and intra-class compactness at different granularities. Experiments on lab-controlled and real-world facial expression dataset show that WS-LGRN achieves state-of-the-art performance, which demonstrates its superiority in FER.

IJCAI Conference 2009 Conference Paper

• Zheng-Jun Zha
• Tao Mei
• Meng Wang
• Zengfu Wang
• Xian-Sheng Hua

Most of the existing metric learning methods are accomplished by exploiting pairwise constraints over the labeled data and frequently suffer from the insufficiency of training examples. To learn a robust distance metric from few labeled examples, prior knowledge from unlabeled examples as well as the metrics previously derived from auxiliary data sets can be useful. In this paper, we propose to leverage such auxiliary knowledge to assist distance metric learning, which is formulated following the regularized loss minimization principle. Two algorithms are derived on the basis of manifold regularization and log-determinant divergence regularization technique, respectively, which can simultaneously exploit label information (i. e. , the pairwise constraints over labeled data), unlabeled examples, and the metrics derived from auxiliary data sets. The proposed methods directly manipulate the auxiliary metrics and require no raw examples from the auxiliary data sets, which make them efficient and flexible. We conduct extensive evaluations to compare our approaches with a number of competing approaches on face recognition task. The experimental results show that our approaches can derive reliable distance metrics from limited training examples and thus are superior in terms of accuracy and labeling efforts.