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


Shuohao Shi has a total of 9 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    WRA-FPN: Weak-Response-Aware Feature Pyramid Network for Tiny Object Detection in Remote Sensing Images

    by Jiyun Zhang, Qiang Fang, Yujun Zeng, Shuohao Shi, Xin Xu

    Tiny object detection remains challenging because tiny instances often occupy only a few pixels, yielding weak, sparse, and easily degraded feature responses. Feature Pyramid Networks (FPNs) are widely used to construct multi-scale feature representations, but they still face two key limitations in tiny object detection. First, conventional FPNs do not explicitly enhance weak target responses. Although existing methods leverage frequency-domain information to strengthen structural details, high-frequency responses can originate from both target boundaries and complex background textures, making indiscriminate enhancement vulnerable to background interference. Second, sparse evidence from tiny objects is highly sensitive to minor spatial deviations during cross-scale propagation and can be diluted by background-dominated feature aggregation. To address these challenges, we propose a Weak-Response-Aware Feature Pyramid Network (WRA-FPN) comprising two complementary components. The Selective Wavelet Enhancement (SWE) module uses scale-adaptive Mexican-hat wavelets to extract localized high-frequency responses and fuses them with input features and high-level semantic features to generate spatial selection weights that enhance target-relevant cues while suppressing background interference. A target–background response constraint further improves feature discriminability. The Adaptive Local–Global Alignment (ALGA) module combines continuous local cross-scale interactions with region-level global dependencies to improve spatial correspondence and preserve weak target evidence, while adaptively fusing local and global information. Extensive experiments on AI-TOD, VisDrone2019, and MS COCO 2017, using multiple representative baseline detectors including RetinaNet, Faster R-CNN, Cascade R-CNN, and DetectoRS, demonstrate improved detection performance of WRA-FPN. On AI-TOD, WRA-FPN improves AP by 4.4% and 2.4% over the corresponding RetinaNet and DetectoRS baselines, respectively.

  • VALA: Virtual anchor-guided label assignment for tiny object detection

    by Jiyun Zhang, Qiang Fang, Shuohao Shi, Yujun Zeng, Xin Xu

  • Open Access

    Dynamic Adaptive Label Assignment for Tiny Object Detection in Remote Sensing Images

    by Shuohao Shi, Qiang Fang, Xin Xu

    With the development of unmanned aerial vehicle and satellite technology, the application of tiny object detection in remote sensing images is becoming increasingly widespread. Although significant progress has been made in the accuracy and speed of object detection in recent years, performance declines sharply when general object detectors are applied to tiny objects; one of the main reasons is unsuitable label assignment strategy. Traditional label assignment strategies often rely on fixed thresholds, leading to mismatches between the number of positive samples and object areas. Additionally, most improved methods require setting one or more hyperparameters. In this paper, we propose a dynamic adaptive label assignment strategy (DALA) comprising three modules. First, we calculate the similarity distance to comprehensively evaluate the matching degree between anchors and each ground truth. Then, we use the ratio‐based label assignment strategy to select an appropriate number of positive samples for each object. Finally, we introduce dynamic weighting loss during training to ensure the model pays more attention to tiny objects. Our three modules automatically adapt to different datasets and detectors without any manual hyperparameter settings. Extensive experiments on four widely used datasets demonstrate the excellent performance of our proposed method. Our code will be released soon.


  • Multiscale Gaussian Attention Mechanism for Tiny-Object Detection in Remote Sensing Images

    by Shuohao Shi, Qiang Fang, Xin Xu, Dezun Dong

    Tiny object detection is increasingly crucial in the fields such as remote sensing, traffic monitoring, and robotics. Inspired by human visual perception, attention mechanism has become a widely used method for enhancing object detection performance. While existing attention mechanisms have significantly advanced general object detection performance, they often fall short in adapting to the characteristics in tiny object datasets, including huge object size variations and concentrated distributions. In detailed, most current attention mechanisms rely on convolutional or linear layers with fixed receptive fields to compute attention vectors. Some methods attempt to enlarge the receptive fields by using multiscale structures, but they often simply sum feature maps, leading to information interference and increased computational costs. To address these issues, we propose a novel Multiscale Gaussian Attention Mechanism (MGAM). This mechanism integrates multiscale receptive fields with dynamic feature weighting and a Gaussian attention module, replacing traditional convolutional layers to reduce training and inference overhead. In additional, our mechanism can be easily embedded into various detectors without any hyperparameters. Extensive experiments on six object detection datasets demonstrate the effectiveness and robustness of our method. Code is available at: https://github.com/cszzshi/MGAM.


  • Open Access

    Joint Optimization Loss Function for Tiny Object Detection in Remote Sensing Images

    by Shuohao Shi, Qiang Fang, Xin Xu

    Tiny object detection remains a formidable challenge in the field of computer vision. There are many factors that influence tiny object detection performance. In this paper, we focus primarily on the following two aspects. First, due to diminutive size and inappropriate label assignment strategy, tiny objects yield significantly fewer positive samples than larger objects, resulting in weakened supervisory signals during backpropagation and model training. Second, most existing detectors directly combine the classification loss and bounding box regression loss during training. Some improvement methods focus exclusively on either classification or localization, leading to potential discrepancies in which predictions exhibit precise localization but incorrect classifications or accurate classifications with imprecise localization. To address these issues, we propose a novel Joint Optimization Loss (JOL) that dynamically assigns optimal weights to each training sample, enabling joint optimization of both the classification and regression losses. Notably, JOL integrates seamlessly with most mainstream detectors and loss functions without requiring alterations to network architectures. Extensive experiments conducted on five benchmark datasets demonstrate the superior performance of our approach, achieving AP improvements of 1.7 and 1.5 points on the AI-TOD and SODA-D datasets, respectively, compared to the state-of-the-art method.


  • Similarity Distance-Based Label Assignment for Tiny Object Detection

    by Shuohao Shi, Qiang Fang, Xin Xu, Tong Zhao

    Tiny object detection is becoming one of the most challenging tasks in computer vision because of the limited object size and lack of information. The label assignment strategy is a key factor affecting the accuracy of object detection. Although there are some effective label assignment strategies for tiny objects, most of them focus on reducing the sensitivity to the bounding boxes to increase the number of positive samples and have some fixed hyperparameters need to set. However, more positive samples may not necessarily lead to better detection results, in fact, excessive positive samples may lead to more false positives. In this paper, we introduce a simple but effective strategy named the Similarity Distance (SimD) to evaluate the similarity between bounding boxes. This proposed strategy not only considers both location and shape similarity but also learns hyperparameters adaptively, ensuring that it can adapt to different datasets and various object sizes in a dataset. Our approach can be simply applied in common anchor-based detectors in place of the IoU for label assignment and Non Maximum Suppression (NMS). Extensive experiments on four mainstream tiny object detection datasets demonstrate superior performance of our method, especially, 1.8 AP points and 4.1 AP points of very tiny higher than the state-of-the-art competitors on AI-TOD. Code is available at: https://github.com/cszzshi/simd.