GZGuoyi Zhang

Guoyi Zhang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Guoyi Zhang has a total of 17 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Robust Non-Contact Vision-Based Measurement of Astronaut Body Keypoint Locations

    by Jingjing Gao, Lejia Ye, Bin Wu, Guangsheng Xu, Guoyi Zhang, Xiangpeng Xu, Xiaohu Zhang

    In crewed space missions, accurate acquisition of astronauts’ body motion states is important for intravehicular and extravehicular operation monitoring, task performance assessment, and mission safety assurance. As fundamental image-based measurements for subsequent three-dimensional pose reconstruction, accurate 2D body keypoint locations are critical to the reliability of downstream 3D motion estimation. However, non-contact keypoint measurement from images remains challenging in spacesuit environments because bulky suit structures, weak joint appearance, limited texture, strong reflectance, occlusion, and large viewpoint variations substantially degrade localization accuracy. In addition, existing spacesuit datasets are limited in scale and mainly cover relatively simple scenarios, restricting systematic evaluation under challenging conditions. To address these limitations, we construct CSPose, a challenging spacesuit keypoint dataset containing over 1,000 images from diverse astronaut task scenarios. We further propose MFRNet, a robust vision-based keypoint measurement framework based on Segment Anything Model 2 (SAM2). MFRNet employs lightweight task adaptation to capture spacesuit-specific visual characteristics and introduces multi-level feature fusion and multi-resolution refinement to preserve spatial details, strengthen structural representation, and improve fine-grained keypoint localization. Experimental results show that MFRNet outperforms representative methods on both CSPose and the public CrowdPose benchmark. With the Hiera-L backbone, MFRNet achieves 85.9 AP on CSPose and 70.5 AP on CrowdPose, demonstrating strong robustness and generalization for non-contact astronaut body keypoint measurement.

  • PanoAir: A Panoramic Visual–Inertial SLAM With Cross-Time Real-World UAV Dataset

    by Yiyang Wu, Xiaohu Zhang, Yanjin Du, Tongsu Zhang, Chujun Li, Siyang Chen, Guoyi Zhang, Xiangpeng Xu

  • Open Access

    You Only Look Omni Gradient Backpropagation for Moving Infrared Small Target Detection

    by Guoyi Zhang, Guangsheng Xu, Siyang Chen, Han Wang, Xiaohu Zhang

    Moving infrared small target detection is a key component of infrared search and tracking systems, yet it remains extremely challenging due to low signal-to-clutter ratios, severe target-background imbalance, and weak discriminative features. Existing deep learning methods primarily focus on spatio-temporal feature aggregation, but their gains are limited, revealing that the fundamental bottleneck lies in ambiguous per-frame feature representations rather than spatio-temporal modeling itself. Motivated by this insight, we propose BP-FPN, a backpropagation-driven feature pyramid architecture that fundamentally rethinks feature learning for small target. BP-FPN introduces Gradient-Isolated Low-Level Shortcut (GILS) to efficiently incorporate fine-grained target details without inducing shortcut learning, and Directional Gradient Regularization (DGR) to enforce hierarchical feature consistency during backpropagation. The design is theoretically grounded, introduces negligible computational overhead, and can be seamlessly integrated into existing frameworks. Extensive experiments on multiple public datasets show that BP-FPN consistently establishes new state-of-the-art performance. To the best of our knowledge, it is the first FPN designed for this task entirely from the backpropagation perspective.

  • It’s Not the Target, It’s the Background: Rethinking Infrared Small-Target Detection via Deep Patch-Free Low-Rank Representations

    by Guoyi Zhang, Guangsheng Xu, Siyang Chen, Han Wang, Xiaohu Zhang

    Infrared small-target detection (IRSTD) remains a long-standing challenge in complex backgrounds due to low signal-to-clutter ratios (SCRs), diverse target morphologies, and the absence of distinctive visual cues. While recent deep learning approaches aim to learn discriminative representations, the intrinsic variability and weak priors of small targets often lead to unstable performance. In this article, we propose a novel end-to-end IRSTD framework, termed low-rank representation network (LRRNet), which leverages the low-rank property of infrared image backgrounds. Inspired by the physical compressibility of cluttered scenes, our approach adopts a compression–reconstruction–subtraction (CRS) paradigm to directly model structure-aware low-rank background representations in the image domain, without relying on patch-based processing or explicit matrix decomposition. To the best of our knowledge, this is the first work to directly learn low-rank background structures using deep neural networks in an end-to-end manner. Extensive experiments on multiple public datasets demonstrate that LRRNet outperforms 38 state-of-the-art methods in terms of detection accuracy, robustness, and computational efficiency. Remarkably, it achieves real-time performance with an average speed of 82.34 FPS. Evaluations on the challenging NoisySIRST dataset further confirm the model’s resilience to sensor noise. The source code will be made publicly available upon acceptance.


  • Open Access

    Beyond Motion Cues and Structural Sparsity: Revisiting Small Moving Target Detection

    by Guoyi Zhang, Siyang Chen, Guangsheng Xu, Zhihua Shen, Han Wang, Xiaohu Zhang

    Small moving target detection is crucial for many defense applications but remains highly challenging due to low signal-to-noise ratios, ambiguous visual cues, and cluttered backgrounds. In this work, we propose a novel deep learning framework that differs fundamentally from existing approaches, which often rely on target-specific features or motion cues and tend to lack robustness in complex environments. Our key insight is that small target detection and background discrimination are inherently coupled, even cluttered video backgrounds often exhibit strong low-rank structures that can serve as stable priors for detection. We reformulate the task as a tensor-based low-rank and sparse decomposition problem and conduct a theoretical analysis of the background, target, and noise components to guide model design. Building on these insights, we introduce TenRPCANet, a deep neural network that requires minimal assumptions about target characteristics. Specifically, we propose a tokenization strategy that implicitly enforces multi-order tensor low-rank priors through a self-attention mechanism. This mechanism captures both local and non-local self-similarity to model the low-rank background without relying on explicit iterative optimization. In addition, inspired by the sparse component update in tensor RPCA, we design a feature refinement module to enhance target saliency. The proposed method achieves state-of-the-art performance on two highly distinct and challenging tasks: multi-frame infrared small target detection and space object detection. These results demonstrate both the effectiveness and the generalizability of our approach.

  • Learning Dynamic Local Context Representations for Infrared Small Target Detection

    by Guoyi Zhang, Guangsheng Xu, Siyang Chen, Han Wang, Xiaohu Zhang

    Infrared small target detection (ISTD) is challenging due to complex backgrounds, low signal-to-clutter ratios (SCRs), and varying target sizes and shapes. Effective detection relies on capturing local contextual information at the appropriate scale. However, small-kernel CNNs have limited receptive fields, leading to false alarms, while transformer models, with global receptive fields, often treat small targets as noise, resulting in miss-detections. Hybrid models struggle to bridge the semantic gap between convolutional neural networks (CNNs) and transformers, causing high complexity. To address these challenges, we propose learning dynamic local context representations network (LCRNet), a novel method that learns dynamic local context representations for ISTD. The model consists of three components: 1) coarse-to-fine convolution block (C2FBlock), inspired by partial differential equation (PDE) solvers, for efficient small target information capture; 2) dynamic local context attention (DLC-Attention), a large-kernel attention mechanism that dynamically builds context and reduces feature redundancy; and 3) hierarchical large-kernel convolution (HLKConv), a hierarchical convolution operator based on large-kernel decomposition that preserves sparsity and mitigates the drawbacks of dilated convolutions. Despite its simplicity, with only 1.65 M parameters, LCRNet achieves state-of-the-art (SOTA) performance. Experiments on multiple datasets, comparing LCRNet with 33 SOTA methods, demonstrate its superior performance and efficiency. Specifically, our method achieves an nIoU exceeding 70% for small object segmentation on the highly challenging IRSTD-1k dataset. Website: https://github.com/HaLongbao/LCRNet/


  • Optimizing Local-Global Dependencies for Accurate 3D Human Pose Estimation

    by Guangsheng Xu, Guoyi Zhang, Lejia Ye, Shuwei Gan, Xiaohu Zhang, Xia Yang

    Transformer-based methods have recently achieved significant success in 3D human pose estimation, owing to their strong ability to model long-range dependencies. However, relying solely on the global attention mechanism is insufficient for capturing the fine-grained local details, which are crucial for accurate pose estimation. Existing local feature extraction networks, such as Graph Convolutional Networks, often suffer from over-smoothing, while small-kernel CNNs have limited receptive fields and are highly sensitive to 2D pose errors. These limitations constrain the full potential of data-driven approaches. To address this, we propose SSR-STF, a dual-stream model that effectively integrates local features with global dependencies to enhance 3D human pose estimation. Specifically, we introduce SSRFormer, a simple yet effective module that employs the skeleton selective refine attention (SSRA) mechanism, leveraging large kernels to capture fine-grained local dependencies in human pose sequences. This complements the global dependencies modeled by the Transformer, enabling a more comprehensive understanding of human motion. By adaptively fusing these two feature streams, SSR-STF can better learn the underlying structure of human poses, overcoming the limitations of traditional methods in local feature extraction. To the best of our knowledge, this is the first work to explore the application of large kernels in skeleton-based 3D human pose estimation. Extensive experiments on the Human3.6M and MPI-INF-3DHP datasets demonstrate that SSR-STF achieves state-of-the-art performance. Furthermore, the motion representations learned by our model prove effective in downstream tasks such as human mesh recovery. Codes are available at SSR-STF.