HHHai-Miao Hu

Hai-Miao Hu

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

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


Hai-Miao Hu has a total of 88 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Fantasy Anything: A Journey of Outlier Imagination Across Unknown Spaces

    by Ruifan Zhang, Yibo Zhou, Xiaokang Zhang, Hai-Miao Hu

  • VTVC: Vignetting Correction Based on Variant Transformer and Spatial Self Attention

    by Zhuang He, Haoxin Hu, Haowen Zheng, Hai-Miao Hu

    Vignetting correction is an essential part for image signal processor. It is important for imaging systems to get high quality images. Traditional vignetting correction methods are almost based on traditional hand-crafted features, which usually perform poorly in complex scenario. Recently, deep learningbased vignetting correction methods have been widely studied. However, the receptive fields of these methods are relatively small, which is difficult for them to estimate the overall vignetting of the image accurately. In this paper, we proposed a deep learning-based vignetting correction model which includes three major components: variant transformer structure, multiple layer upsampling and spatial self attention. The variant transformer structure is construct by replacing the multi-layer perceptron of traditional transformer with a structure extractor module, so that to improve the receptive fields for vignetting estimation. The multiple layer upsampling and spatial self attention are used to improve the resolution and preserve the image details, thereby improving the accuracy of vignetting correction. Additionally, due to the lack of vignetting correction dataset, we introduced a method to get vignetting images through capturing the real world, and construct the first thousand-level vignetting correction dataset. Besides, we also proposed a data augmentation method based on the optical principle of vignetting. Through training on the large amount of real-world vignetting dataset, the proposed vignetting correction network can be trained accurately. Many experiments have been done on the conducted dataset. Through the comparison with the state-of-the-art algorithms, it has been verified that the proposed vignetting correction network demonstrates the best performance.

  • A Multi-Category Anomaly Editing Network With Correlation Exploration and Voxel-Level Attention for Unsupervised Surface Anomaly Detection

    by Ruifan Zhang, Hai-Miao Hu

    Developing a unified model for surface anomaly detection remains challenging due to significant variations across product categories. Recent feature editing methods, as a branch of image reconstruction, mitigate the over-generalization of auto-encoders that leads to accurate anomaly reconstruction. However, these methods are only suited for texture-category products and have significant limitations in being generalized to other categories. In this article, we propose a multi-category anomaly editing network with a dual-branch training approach: one branch processes defect-free images (normal branch), while the other handles synthetic anomaly images (anomaly branch). Specifically, the paired samples are first fed into the multi-category anomaly feature editing based auto-encoder (MCAFE-AE) to perform image reconstruction and inpainting. In the normal branch, we propose a dual-entropy constrained deep embedded clustering module (DEC-DECM) to promote a more compact and orderly distribution of normal latent features, while avoiding trivial clustering solutions. Based on the clustering results, we further design a patch-based adaptive thresholding (PAT) strategy to adaptively calculate the threshold representing the central boundary of the cluster center for each local patch, thereby enabling the model to detect anomalies. Then, in the anomaly branch, we propose a multi-category anomaly feature editing module (MCAFEM) to identify anomalies in synthetic images and apply a category-oriented feature editing strategy to transform detected anomaly features into normal ones, thereby suppressing the reconstruction of anomalies. After completing the image reconstruction and inpainting, the input images from both branches and their respective output images are concatenated and fed into the correlation exploration and voxel-level attention based prediction network (CEVA-Net) for anomaly segmentation. The network is integrated with our proposed correlation-dependency exploration and voxel-level attention refinement module (CDE-VARM) and generates precise anomaly maps under the guidance of the bidirectional-path feature fusion (BPFF) and deep supervised learning (DSL). Extensive experiments on three datasets show that our method achieves state-of-the-art performance.


  • Towards Multi-Source Illumination Color Constancy Through Physics-Based Rendering and Spectral Power Distribution Embedding

    by Xinhui Xue, Hai-Miao Hu, Zhuang He, Haowen Zheng

    Color constancy seeks to keep the perceived color of objects consistent under varying illumination conditions. However, existing methods often rely on restrictive prior assumptions or suffer from limited generalization capability, posing significant challenges in complex scenes with multiple light sources. In this paper, we propose a neural network-enhanced, physicsbased approach to multi-illuminant color constancy that leverages spectral imaging—highly sensitive to illumination variation. First, we analyze the physical image-formation process under mixed lighting and introduce a master–subordinate illumination model, extending conventional correlated-color-temperature reillumination techniques. Our neural network framework ex-plicitly models the correlation between narrow-band spectral reflectance and the spectral power distribution (SPD) of the illumination, enabling accurate recovery of the scene light's full SPD. Using this model, we fuse RGB images with the estimated illumination spectra to predict illuminant chromaticity precisely, then correct image colors to a standard reference light. Extensive experiments on synthetic multi–color-temperature datasets and real-world spectral datasets demonstrate that our neural network-based method achieves state-of-the-art accuracy in spectral estimation and color-constancy correction.


  • Vignetting Correction Through Color-Intensity Map Entropy Optimization

    by Zhuang He, Hai-Miao Hu, Likun Gao, Haoxin Hu, Xinhui Xue, Zhenglin Tang, Difeng Zhu, Haowen Zheng, Chongze Wang

    Vignetting correction is an essential process of image signal processing. It is an important part for obtaining high-quality images, but the research in this field has not been fully emphasized. The mainstream methods are based on calibration which processes are complex. And many methods get low accuracy and poor robustness in practical. In this paper, we analyzed the optical principle of vignetting and its influence on the image. Then, we proposed an algorithm based on color-intensity map entropy optimization to correct image vignetting. Moreover, because of the lack of dataset of vignetting, we proposed a method for constructing vignetting image dataset through capturing the real scenes. Compared with the dataset generated through simulation, our dataset is more authentic and reliable. Many experiments have been carried out on this dataset, and the results proved that the proposed algorithm achieved the best performance.


  • Open Access

    Consistency Preserved Nonuniform Scattering Removal for Wide-Swath Remote Sensing Images

    by Yurui Liu, Huaizhuo Liu, Ruifan Zhang, Zhenglin Tang, Hai-Miao Hu

    Removal of haze and thin clouds in remote sensing images has been extensively studied, leading to the development of effective methods. However, when applied to wide-swath images, conventional dehazing approaches often generate patch-level inconsistencies during combination. To address the issue, this paper presents a consistency preserved scattering model designed with multiple features to remove haze and thin cloud effects. Specifically, three modules named ESD, ILM and SPC are introduced to improve the basic prior-based dehazing method. ESD focuses on cloud area exclusion to ensure accurate estimation with the scattering distribution map. ILM is based on the linear relationship with scattering distribution and illumination, aiming to smooth the global illumination. SPC addresses the spectral adaptation issues in wide-swath scenes. Experimental results indicate that, compared to state-of-the-art methods, the proposed model and method effectively restores textural details in remote sensing imagery affected by heterogeneous haze, while resolving cross-patch inconsistency to enhance overall image coherence and interpretability.


  • CGATracker: Correlation-Aware Graph Alignment for Referring Multi-Object Tracking

    by Siping Zhuang, Guangyao Li, Qiangqiang Wu, Yang Lu, Hai-Miao Hu, Hanzi Wang

    Referring multi-object tracking (RMOT) aims to identify specific targets based on sentence descriptions. To enhance multi-modal learning, previous works typically relied on a simple fusion module at early or late stages. However, those methods frequently underutilize textual semantics and struggle to model the relationships between region-level features and word-level features. To address these limitations, we propose CGATracker, a correlation-aware graph alignment method for RMOT, which facilitates precise relationship modeling through relational scoring. Specifically, we design a Language-driven Relational Alignment (LRA) module, which establishes two connection graphs to generate positive and negative samples for the visual-textual alignment. Additionally, to effectively leverage referring information, we introduce a Semantic Clarify Booster (SCBooster) module based on a semantic infusion mechanism and a bias-aware verification mechanism for interactions with different modalities. Moreover, by designing a Multi-level Cross-modal Fusion (MCF) module, our method aggregates contextual features at multiple depths to enable the creation of the enriched correlation-aware graph. Extensive experiments conducted on the Refer-KITTI and Refer-KITTI-V2 datasets demonstrate the effectiveness of CGATracker.


  • Open Access

    PEIE: Physics Embedded Illumination Estimation for Adaptive Dehazing

    by Huaizhuo Liu, Hai-Miao Hu, Yonglong Jiang, Yurui Liu

    Deep learning-based methods have made significant progress in image dehazing. However, these methods often falter when applied to real-world hazy images, primarily due to the scarcity of paired real-world data and the limitations of current dehazing feature extractors. Toward these issues, we introduce a novel Physics Embedded Illumination Estimation (PEIE) method for adaptive real-world dehazing. Specifically, (1) we identify the limitations of the widely used Atmospheric Scattering Model and propose a new physical model, the Illumination-Adaptive Scattering Model (IASM), for more accurate illumination representation in hazy imaging; (2) we develop a robust data synthesis pipeline that leverages the physics embedded illumination estimation to generate realistic hazy images; and (3) we design an Illumination-Adaptive Dehazing Unit (IDU) to extract dehazing features consistent with our proposed IASM in the latent space. By integrating the IDU into a U-Net architecture to create IADNet, we achieve significant improvements in dehazing performance through end-to-end training on synthetic data. Extensive experiments validate the superior performance of our PEIE method, significantly surpassing the state-of-the-arts in real-world dehazing.


  • A Solution to Co-occurrence Bias in Pedestrian Attribute Recognition: Theory, Algorithms, and Improvements

    by Yibo Zhou, Hai-Miao Hu, Jinzuo Yu, Haotian Wu, Shiliang Pu, Hanzi Wang