arXiv subjects

Hongtao Yang

Publications and source records attributed to Hongtao Yang.

7 recordsLinked to original sources

arXiv2026

Dual-branch Distilled Transformer for Efficient Asymmetric UAV Tracking

Given the real-time demands of UAV tracking, many methods simplify the backbone to reduce computation, but this often weakens feature representation and degrades performance in complex scenarios. To alleviate this issue, we propose EATrack, an efficient and asymmetric UAV tracking framework centered around a teacher-guided dual-branch distillation strategy that enhances the feature expressiveness of the lightweight student model. Specifically, EATrack investigates two complementary perspectives of knowledge transfer: spatially focused feature-level distillation that compensates for weakened representations by guiding the student to learn strong target representations, and prediction-level distillation that enhances spatial localization by learning the teacher's capability for accurate target localization. Furthermore, to enhance robustness against appearance variations, we introduce a fine-grained target-aware distillation strategy that selectively transfers the teacher's target modeling capacity to the student. A temporal adaptation module is incorporated at inference to enhance robustness over time. Experiments on five UAV benchmarks demonstrate that EATrack achieves a favorable balance between accuracy and speed. Code: https://github.com/GXNU-ZhongLab/EATrack

cs.CV↗

arXiv2025

Robust RGB-T Tracking via Learnable Visual Fourier Prompt Fine-tuning and Modality Fusion Prompt Generation

Recently, visual prompt tuning is introduced to RGB-Thermal (RGB-T) tracking as a parameter-efficient finetuning (PEFT) method. However, these PEFT-based RGB-T tracking methods typically rely solely on spatial domain information as prompts for feature extraction. As a result, they often fail to achieve optimal performance by overlooking the crucial role of frequency-domain information in prompt learning. To address this issue, we propose an efficient Visual Fourier Prompt Tracking (named VFPTrack) method to learn modality-related prompts via Fast Fourier Transform (FFT). Our method consists of symmetric feature extraction encoder with shared parameters, visual fourier prompts, and Modality Fusion Prompt Generator that generates bidirectional interaction prompts through multi-modal feature fusion. Specifically, we first use a frozen feature extraction encoder to extract RGB and thermal infrared (TIR) modality features. Then, we combine the visual prompts in the spatial domain with the frequency domain prompts obtained from the FFT, which allows for the full extraction and understanding of modality features from different domain information. Finally, unlike previous fusion methods, the modality fusion prompt generation module we use combines features from different modalities to generate a fused modality prompt. This modality prompt is interacted with each individual modality to fully enable feature interaction across different modalities. Extensive experiments conducted on three popular RGB-T tracking benchmarks show that our method demonstrates outstanding performance.

cs.CV↗

arXiv2024

Machine learning disentangles bias causes of shortwave cloud radiative effect in a climate model

Large bias exists in shortwave cloud radiative effect (SWCRE) of general circulation models (GCMs), attributed mainly to the combined effect of cloud fraction and water contents, whose representations in models remain challenging. Here we show an effective machine-learning approach to dissect the individual bias of relevant cloud parameters determining SWCRE. A surrogate model for calculating SWCRE was developed based on random forest using observations and FGOALS-f3-L simulation data of cloud fraction (CFR), cloud-solar concurrence ratio (CSC), cloud liquid and ice water paths (LWP and IWP), TOA upward clear-sky solar flux (SUC), and solar zenith angle. The model, which achieves high determination coefficient > 0.96 in the validation phase, was then used to quantify SWCRE bias associated with these parameters following the partial radiation perturbation method. The global-mean SWCRE bias (in W m-2) is contributed by CFR (+5.11), LWP (-6.58), IWP (-1.67), and CSC (+4.38), while SUC plays a minor role; the large CSC contribution highlights the importance of cloud diurnal variation. Regionally, the relative importance varies according to climate regimes. In Tropics, overestimated LWP and IWP exist over lands, while oceans exhibit underestimated CFR and CSC. In contrast, the extratropical lands and oceans have, respectively, too-small CSC and the 'too few, too bright' low-level clouds. We thus suggest that machine learning, in addition for developing GCM physical parameterizations, can also be utilized for diagnosing and understanding complex cloud-climate interactions.

physics.ao-ph↗

arXiv2023

Infrared Imaging of Magnetic Octupole Domains in Non-collinear Antiferromagnets

Magnetic structure plays a pivotal role in the functionality of antiferromagnets (AFMs), which not only can be employed to encode digital data but also yields novel phenomena. Despite its growing significance, visualizing the antiferromagnetic domain structure remains a challenge, particularly for non-collinear AFMs. Currently, the observation of magnetic domains in non-collinear antiferromagnetic materials is feasible only in Mn$_{3}$Sn, underscoring the limitations of existing techniques that necessitate distinct methods for in-plane and out-of-plane magnetic domain imaging. In this study, we present a versatile method for imaging the antiferromagnetic domain structure in a series of non-collinear antiferromagnetic materials by utilizing the anomalous Ettingshausen effect (AEE), which resolves both the magnetic octupole moments parallel and perpendicular to the sample surface. Temperature modulation due to the AEE originating from different magnetic domains is measured by the lock-in thermography, revealing distinct behaviors of octupole domains in different antiferromagnets. This work delivers an efficient technique for the visualization of magnetic domains in non-collinear AFMs, which enables comprehensive study of the magnetization process at the microscopic level and paves the way for potential advancements in applications.

cond-mat.mtrl-sci↗

arXiv2020

Towards Purely Unsupervised Disentanglement of Appearance and Shape for Person Images Generation

There have been a fairly of research interests in exploring the disentanglement of appearance and shape from human images. Most existing endeavours pursuit this goal by either using training images with annotations or regulating the training process with external clues such as human skeleton, body segmentation or cloth patches etc. In this paper, we aim to address this challenge in a more unsupervised manner---we do not require any annotation nor any external task-specific clues. To this end, we formulate an encoder-decoder-like network to extract both the shape and appearance features from input images at the same time, and train the parameters by three losses: feature adversarial loss, color consistency loss and reconstruction loss. The feature adversarial loss mainly impose little to none mutual information between the extracted shape and appearance features, while the color consistency loss is to encourage the invariance of person appearance conditioned on different shapes. More importantly, our unsupervised (Unsupervised learning has many interpretations in different tasks. To be clear, in this paper, we refer unsupervised learning as learning without task-specific human annotations, pairs or any form of weak supervision.) framework utilizes learned shape features as masks which are applied to the input itself in order to obtain clean appearance features. Without using fixed input human skeleton, our network better preserves the conditional human posture while requiring less supervision. Experimental results on DeepFashion and Market1501 demonstrate that the proposed method achieves clean disentanglement and is able to synthesis novel images of comparable quality with state-of-the-art weakly-supervised or even supervised methods.

cs.CV↗

arXiv2019

Simplified Template Cross Sections -- Stage 1.1 and 1.2

Simplified Template Cross Sections (STXS) have been adopted by the LHC experiments as a common framework for Higgs measurements. Their purpose is to reduce the theoretical uncertainties that are directly folded into the measurements as much as possible, while at the same time allowing for the combination of the measurements between different decay channels as well as between experiments. We report the complete, revised definition of the STXS kinematic bins (stage 1.1 and stage 1.2), which have been used for the measurements by the ATLAS and CMS experiments using the full LHC Run 2 datasets. The main focus is on the four dominant Higgs production processes, namely gluon-fusion, vector-boson fusion, production in association with a vector boson and in association with a $t\bar t$ pair. We also comment briefly on the treatment of other production modes.

hep-ph↗

arXiv2018

Optimizing Floating Locations in Hard Disk Drive by Solving Max-min Optimization

Floating operation is very critical in power management in hard disk drive (HDD), during which no control command is applied to the read/write head but a fixed current to counteract actuator flex bias. External disturbance induced drift of head may result in interference of head and bump on the disk during drifting, leading to consequent scratches and head degradation, which is a severe reliability concern in HDD. This paper proposes a unique systematic methodology to minimize the chances of hitting bump on the disk during drive floating. Essentially, it provides a heuristic solution to a class of max-min optimization problem which achieves desirable trade-off between optimality and computation complexity. Multivariable nonlinear optimization problem of this sort is reduced from NP-hard to an arithmetic problem. Also, worst-case is derived for arbitrary bump locations.

eess.SY↗