Tao Huang

Dr. Tao Huang

orcid

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
22 December 2025 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
YFYuxin Feng
Yuxin Feng
JLJufeng Li
Jufeng Li
Tao HuangTao Huang
Tao Huang
FWFangfang Wu
Fangfang Wu
Yakun JuYakun Ju
Yakun Ju
AKAlex C. Kot
Alex C. Kot
Current deep learning-based methods for remote sensing image dehazing have developed rapidly, yet they still commonly struggle to simultaneously preserve fine texture details and restore accurate colors. The fundamental reason lies in the insufficient modeling of high-frequency information that captures structural details, as well as the lack of effective constraints for color restoration. To address the insufficient modeling of global high-frequency information, we first develop an omni-directional high-frequency feature inpainting mechanism that leverages the wavelet transform to extract multi-directional high-frequency components. While maintaining the advantage of linear complexity, it models global long-range texture dependencies through cross-frequency perception. Then, to further strengthen local high-frequency representation, we design a high-frequency prompt attention module that dynamically injects wavelet-domain optimized high-frequency features as cross-level guidance signals, significantly enhancing the model’s capability in edge sharpness restoration and texture detail reconstruction. Further, to alleviate the problem of inaccurate color restoration, we propose a color contrast loss function based on the HSV color space, which explicitly models the statistical distribution differences of brightness and saturation in hazy regions, guiding the model to generate dehazed images with consistent colors and natural visual appearance. Finally, extensive experiments on multiple benchmark datasets demonstrate that the proposed method outperforms existing approaches in both texture detail restoration and color consistency. Further results and code available at: https://github.com/fyxnl/C4RSD.

26 November 2025 in Information Fusion
Verified
Journal Article
Closed access
JXJunwei Xu
Junwei Xu
ZFZhenxuan Fang
Zhenxuan Fang
MLMengzu Liu
Mengzu Liu
FWFangfang Wu
Fangfang Wu
Tao HuangTao Huang
Tao Huang
GSGuangming Shi
Guangming Shi

14 October 2025 in IEEE Journal of Selected Topics in Signal Processing
Verified
Journal Article
Closed access
JXJunwei Xu
Junwei Xu
MLMengzu Liu
Mengzu Liu
SZShilong Zhang
Shilong Zhang
FWFangfang Wu
Fangfang Wu
Tao HuangTao Huang
Tao Huang
WDWeisheng Dong
Weisheng Dong
Hyperspectral image (HSI) demosaicing plays a critical role in reconstructing full-resolution HSIs from multispectral filter array (MSFA) acquisitions. Although recent deep-learning approaches based on CNNs and Transformers have achieved significant improvements, a systematic understanding of how these architectures behave across different frequency components remains limited. To address this limitation, we perform a comprehensive frequency-aware analysis with wavelet decomposition, systematically revealing that Transformers excel at modeling low-frequency global structures, whereas CNNs are superior for reconstructing high-frequency details within the unique context of HSI demosaicing. Motivated by this principled insight, we propose Waveformer, a novel frequency-aware hybrid framework that explicitly decomposes the HSI into low- and high-frequency components using wavelet transform and assigns their reconstruction tasks to optimally suited architectures: Transformers for low-frequency modeling and CNNs for high-frequency enhancement. Furthermore, to counteract ambiguity in traditional attention mechanisms caused by softmax normalization, we introduce a Self-Calibration Attention mechanism that adaptively calibrates each token's self-identity across spatial and spectral dimensions, thereby enhancing the model's discriminative power and suppressing interactions from irrelevant tokens. Extensive experiments conducted on multiple benchmark datasets validate the effectiveness of our design. Our method consistently outperforms state-of-the-art baselines, demonstrating strong generalization ability and enhanced reconstruction quality across diverse scenes and spectral configurations.

10 June 2025
Verified
Conference Paper
Closed access
ZYZhou Yang
Zhou Yang
MFMingtao Feng
Mingtao Feng
Tao HuangTao Huang
Tao Huang
FWFangfang Wu
Fangfang Wu
WDWeisheng Dong
Weisheng Dong
GSGuangming Shi
Guangming Shi
Recent approaches, such as data augmentation, adversarial training, and transfer learning, have shown potential in addressing the issue of performance degradation caused by distributional shifts. However, they typically demand careful design in terms of data or models and lack awareness of the impact of distributional shifts. In this paper, we observe that classification errors arising from distribution shifts tend to cluster near the true values, suggesting that misclassifications commonly occur in semantically similar, neighboring categories. Furthermore, robust advanced vision foundation models maintain larger inter-class distances while preserving semantic consistency, making them less vulnerable to such shifts. Building on these findings, we propose a new method called GFN (Gain From Neighbors), which uses gradient priors from neighboring classes to perturb input images and incorporates an inter-class distance-weighted loss to improve class separation. This approach encourages the model to learn more resilient features from data prone to errors, enhancing its robustness against shifts in diverse settings. In extensive experiments across various model architectures and benchmark datasets, GFN consistently demonstrated superior performance. For instance, compared to the current state-of-the-art TAPADL method, our approach achieved a higher corruption robustness of 41.4% on ImageNet-C (+2.3%), without requiring additional parameters and using only minimal data.

28 August 2025 in International Journal of Computer Vision
Verified
Journal Article
Closed access
PDF file
ZWZhenyu Wang
Zhenyu Wang
Xuemei XieXuemei Xie
Xuemei Xie
HLHao Luo
Hao Luo
Tao HuangTao Huang
Tao Huang
WDWeisheng Dong
Weisheng Dong
GSGuangming Shi
Guangming Shi

14 May 2026 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
JXJunwei Xu
Junwei Xu
TFTeng Feng
Teng Feng
ZFZhenxuan Fang
Zhenxuan Fang
FWFangfang Wu
Fangfang Wu
LDLe Dong
Le Dong
XLXin Li
Xin Li
s an alternative to acquiring high-resolution hyperspectral images (HR-HSI), Hyperspectral Image Fusion (HIF) aims to recover clean HR-HSIs by fusing degraded low spatial resolution hyperspectral images and high spatial resolution multispectral images. Existing model-guided HIF methods elegantly integrate physical degradation constraints with the powerful learning capabilities of data-driven networks.s an alternative to acquiring high-resolution hyperspectral images (HR-HSI), Hyperspectral Image Fusion (HIF) aims to recover clean HR-HSIs by fusing degraded low spatial resolution hyperspectral images and high spatial resolution multispectral images. Existing model-guided HIF methods elegantly integrate physical degradation constraints with the powerful learning capabilities of data-driven networks.A However, these methods struggle with severe, unseen degradations, as their deep priors are trained on degraded-clean pairs and thus lack knowledge of clean HSI characteristics. To address these issues, we introduce a Vector-Quantized Prior-Guided Network (VPG-Net), which incorporates a novel uncertainty-driven generative model prior alongside a traditional sparse representation prior. Specifically, VPG-Net unfolds the Maximum A Posteriori (MAP) estimation with a sparse representation model into an uncertainty-aware generative prior-guided network implementation. The sparse representation prior is integrated into the MAP framework to improve noise resistance, whose solution is implemented as a trainable neural network. As the core of our contribution, we leverage a high-quality vector-quantized (VQ) prior, which serves as a powerful degradation-free generative prior for the HIF process. Specifically, we pre-train a discrete codebook and encoder from clean HR-HSIs to generate a VQ-prior representation (VQPR), which preserves complete spatial-spectral information. To effectively bridge the gap between degraded inputs and the learned degradation-free codebook, we incorporate a novel uncertainty-driven probabilistic matching strategy that robustly aligns features and prevents artifacts. The learned VQPR is then incorporated into the reconstruction model as dynamic modulation parameters to enhance the fidelity and realism of the reconstructed results, particularly for severely degraded inputs. Extensive experiments on clean and degraded synthetic and real-world datasets demonstrate that our approach outperforms state-of- the-art HIF methods in both quantitative metrics and visual quality.