Panpan Zheng

Panpan Zheng

Auto-generated profile, based on publicly available information

Panpan Zheng

Xinjiang University

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

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

Representative Publications

View All Publications
24 May 2024 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Open access
NHNaiwei He
Naiwei He
LWLiejun Wang
Liejun Wang
PZPanpan Zheng
Panpan Zheng
CZCui Zhang
Cui Zhang
LLLele Li Li
Lele Li Li
Remote sensing image change detection (CD) is an important technology for monitoring ground object change. Although Transformer-based CD methods have been proposed and achieved good results, however, there does exist one open problem: Transformer-based methods are weak for localizing information acquisition, easily ignore detailed information, and are of high computational complexity. Also, the variation of target sizes challenges the generalization of networks. To address these issues, we propose a siamese network named as CBSASNet for remote sensing CD, in which channel bias split attention is employed to recover the information in the change region and the cross-temporal fusion module is utilized to highlight the information of change regions through the optimized single-temporal image features. The experimental result indicates that CBSASNet does not only outperform 16 state-of-the-art works, but also its modules complement each other in the ablation testing.

13 March 2025 in Expert Systems with Applications
Verified
Journal Article
Closed access
YLYuxuan Luo
Yuxuan Luo
YXYongquan Xue
Yongquan Xue
YTYifei Teng
Yifei Teng
LWLiejun Wang
Liejun Wang
PZPanpan Zheng
Panpan Zheng

11 April 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
RZRundong Zhao
Rundong Zhao
PZPanpan Zheng
Panpan Zheng
CZCui Zhang
Cui Zhang
LWLiejun Wang
Liejun Wang
The existing salient object detection in optical remote sensing images methods mostly employ the same strategy to handle features at different levels without fully considering the distinct characteristics these features possess at various levels. This results in the neglect of some high-level semantics and low-level details during the feature extraction process. Furthermore, existing methods often rely on simple convolution operations to construct modules for feature extraction and fusion. Due to the inherent locality of convolution operations, these models are limited in their performance. To address these challenges, we propose a novel progressive complementation network with semantics and details (SDPCNet) consisting of three parts: Deep semantics aggregation module (DSAM), semantics-guided feature complement module (SFCM), and detail feature enhancement module (DFEM). Specifically, the DSAM is applied on the two highest-level features, guided by the global view with global long-range dependencies and local context generated by transformer and dilated convolution. The DSAM deeply delves the semantic information in high-level features to perceive the object positions and alleviate the adverse effects of cluttered backgrounds. The SFCM operates on the intermediate two levels of features, performing global correlation modeling on the aggregated cross-level features. It enhances multiscale semantic information and edge details using multiple sets of dilated convolutions to address the challenges posed by the uncertainty in the size and number of salient objects. The DFEM acts on the lowest two levels of features, enhancing edge details in spatial dimension and emphasizing semantics in different channel dimensions. It is then fused with high-level features to augment feature diversity and reduce the impact of background noise. Extensive experiments conducted on the ORSSD, EORSSD, and ORSI-4199 datasets demonstrate that our proposed SDPCNet outperforms 23 state-of-the-art methods across eight evaluation metrics.

28 October 2024
Verified
Conference Paper
Open access
XFXiang Fang
Xiang Fang
WFWanlong Fang
Wanlong Fang
DLDaizong Liu
Daizong Liu
XQXiaoye Qu
Xiaoye Qu
JDJianfeng Dong
Jianfeng Dong
YCYu Cheng
Yu Cheng
Video Moment Retrieval (VMR) targets to retrieve the specific moment corresponding to a sentence query from an untrimmed video. Although recent respectable works have made remarkable progress in this task, they implicitly are rooted in the closed-set assumption that all the given queries as video-relevant. Given an OOD query in open-set scenarios, they still utilize it for wrong retrieval, which might lead to irrecoverable losses in high-risk scenarios, e.g., criminal activity detection. To this end, we creatively explore a brand-new VMR setting termed Open-Set Video Moment Retrieval (OS-VMR), where we should not only retrieve the precise moments based on ID query, but also reject OOD queries. In this paper, we make the first attempt to step toward OS-VMR and propose a novel model OpenVMR, which first distinguishes ID and OOD queries based on the normalizing flow technology, and then conducts moment retrieval based on ID queries. Specifically, we first learn the ID distribution by constructing a normalizing flow, and assume the ID query distribution obeys the multi-variate Gaussian distribution. Then, we introduce an uncertainty score to search the ID-OOD separating boundary. After that, we refine the ID-OOD boundary by pulling together ID query features. Besides, video-query matching and frame-query matching are designed for coarse-grained and fine-grained cross-modal interaction, respectively. Finally, a positive-unlabeled learning module is introduced for moment retrieval. Experimental results on three VMR datasets show the effectiveness of our OpenVMR.

15 January 2025 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
JLJingming Li
Jingming Li
PZPanpan Zheng
Panpan Zheng
LWLiejun Wang
Liejun Wang
As deep learning demonstrates excellent performance in remote sensing image change detection (CD), early methods that mostly used Convolutional Neural Networks (CNNs) have limitations in the accuracy due to their insufficient global feature representation, an inherent shortcoming of CNNs. The lack of global feature can lead to notable issues, such as the inability to detect small targets and loss of edge information. In recent years, vision transformers (ViTs) have been employed in CD owing to their powerful global feature representation capabilities. However, pure transformer methods lack effective local feature extraction, which also restricts the performance of CD, while the original transformer models require a large amount of computing resources. To address these issues and improve CD performance, we propose a Lightweight Transformer-based Multi-scale Feature Fusion network (LTMFFNet). By integrating CNN structures both before and after the multi-head self-attention in each layer of the main backbone, we enhance the encoder's local feature extraction ability and reduce the computational complexity through convolution and linear operations. For the siamese encoding outputs at different scales, we design two distinct fusion modules based on depth- wise convolution for bitemporal information fusion in deep layers and shallow layers, respectively. Our model employs a multi-layer cascaded structure with a deep supervision strategy applied to multiple outputs. Experiments on four public CD datasets demonstrate that our network achieves better performance while maintaining relatively smaller computational complexity compared to other state-of-the-art methods for CD.

24 November 2025 in Knowledge-Based Systems
Verified
Journal Article
Closed access
TYTong Yumei
Tong Yumei
PZPanpan Zheng
Panpan Zheng
WTWenbin Tang
Wenbin Tang
LWLiejun Wang
Liejun Wang