Bo Peng

Dr. Bo Peng


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
11 April 2025 in Proceedings of the AAAI Conference on Artificial Intelligence
Verified
Journal Article
Open access
PDF file
MYMufeng Yao
Mufeng Yao
JPJinlong Peng
Jinlong Peng
QHQingdong He
Qingdong He
BPBo Peng
Bo Peng
HCHao Chen
Hao Chen
Jón Atli BenediktssonJón Atli Benediktsson
Jón Atli Benediktsson
Multiple object tracking (MOT) from unmanned aerial vehicle (UAV) platforms requires efficient motion modeling. This is because UAV-MOT faces both local object motion and global camera motion. Motion blur also increases the difficulty of detecting large moving objects. Previous UAV motion modeling approaches either focus only on local motion or ignore motion blurring effects, thus limiting their tracking performance and speed. To address these issues, we propose the Motion Mamba Module, which explores both local and global motion features through cross-correlation and bi-directional Mamba Modules for better motion modeling. To address the detection difficulties caused by motion blur, we also design motion margin loss to effectively improve the detection accuracy of motion blurred objects. Based on the Motion Mamba module and motion margin loss, our proposed MM-Tracker surpasses the state-of-the-art in two widely open-source UAV-MOT datasets.

6 March 2026 in Pattern Recognition
Verified
Journal Article
Closed access
WZWenbing Zhu
Wenbing Zhu
CWChengjie Wang
Chengjie Wang
BGBin-Bin Gao
Bin-Bin Gao
JZJiangning Zhang
Jiangning Zhang
GJGuannan Jiang
Guannan Jiang
LMLizhuang Ma
Lizhuang Ma

Verified
Journal Article
Open access
PDF file
BPBo Peng
Bo Peng
Zhonghua HongZhonghua Hong
Zhonghua Hong
GWGuansuo Wang
Guansuo Wang
Accurate multi-day sea surface temperature (SST) prediction at sub-mesoscale resolution is challenging due to nonlinear ocean dynamics, heterogeneous multi-source observations, and error accumulation during autoregressive rollout. This paper proposes a hybrid Mamba–ConvLSTM framework that combines recurrent local spatiotemporal encoding with selective state-space long-range spatial modeling. The ConvLSTM branch captures local spatial patterns and short-range temporal dependencies through convolutional gating, while the Mamba branch captures long-range spatial dependencies across each frame through cross-direction window scanning and maintains temporal coherence via persistent hidden states across successive time steps. A physically informed preprocessing stage aligns 0.083° reanalysis variables to the 0.05° OSTIA target grid via a Grow-and-Cut strategy and extracts gradient-based advection and diffusion proxy features under boundary-aware finite differencing. During autoregressive rollout, auxiliary variables are held at their last observed values and physical proxies are recomputed from the predicted SST, following a clearly specified protocol. Experiments on a South China Sea benchmark compare the proposed model against nine baselines—including persistence, daily climatology, ConvLSTM, PredRNN, ConvGRU, TCTN, PANN, Swin-UNet, and ViT-ST—under an identical data-split, normalization, and rollout protocol. Evaluation with RMSE, MAE, SSIM, R2, and anomaly correlation coefficient (ACC) shows that the proposed model achieves a 10-day average RMSE of 0.512 °C, outperforming the strongest learning-based baseline ViT-ST by 5.0% and the persistence forecast by 21.0%. Ablation studies, sensitivity analyses, seasonal evaluation, and statistical significance testing verify the contribution of each component and the robustness of the results.

19 April 2025 in arXiv
Verified
Preprint
Open access
PDF file
WZWenbing Zhu
Wenbing Zhu
LWLidong Wang
Lidong Wang
ZZZiqing Zhou
Ziqing Zhou
CWChengjie Wang
Chengjie Wang
YPYurui Pan
Yurui Pan
LMLizhuang Ma
Lizhuang Ma
The increasing complexity of industrial anomaly detection (IAD) has positioned multimodal detection methods as a focal area of machine vision research. However, dedicated multimodal datasets specifically tailored for IAD remain limited. Pioneering datasets like MVTec 3D have laid essential groundwork in multimodal IAD by incorporating RGB+3D data, but still face challenges in bridging the gap with real industrial environments due to limitations in scale and resolution. To address these challenges, we introduce Real-IAD D3, a high-precision multimodal dataset that uniquely incorporates an additional pseudo3D modality generated through photometric stereo, alongside high-resolution RGB images and micrometer-level 3D point clouds. Real-IAD D3 features finer defects, diverse anomalies, and greater scale across 20 categories, providing a challenging benchmark for multimodal IAD Additionally, we introduce an effective approach that integrates RGB, point cloud, and pseudo-3D depth information to leverage the complementary strengths of each modality, enhancing detection performance. Our experiments highlight the importance of these modalities in boosting detection robustness and overall IAD performance. The dataset and code are publicly accessible for research purposes at https://realiad4ad.github.io/Real-IAD D3

15 December 2025 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
YZYun Zhang
Yun Zhang
TWTianyue Wen
Tianyue Wen
SYShuhu Yang
Shuhu Yang
QSQingjing Shi
Qingjing Shi
QQQifeng Qian
Qifeng Qian
Zhonghua HongZhonghua Hong
Zhonghua Hong
The development of GNSS-R ocean altimetry technology has significantly advanced global sea surface height (SSH) monitoring. However, single-satellite systems face inherent limitations: their spatial coverage is constrained by orbital patterns and revisit cycles, making them insufficient for high-precision, high spatiotemporal resolution global SSH monitoring. This paper innovatively proposes a deep learning model for multi-source spaceborne GNSS-R fusion-based SSH retrieval. The model corrects low-accuracy single-source inversion results and integrates inversion outputs from four GNSS-R satellite systems: FY-3E, FY-3G, CYGNSS, and Tianmu-1. This approach reduces errors caused by single sources and improves the coverage area of SSH measurements. Experiments were conducted globally between 60°N and 60°S, where the higher-precision FY-3E SSH data were used to correct the retrieval results of FY-3G, CYGNSS, and Tianmu-1. Under relatively lenient data quality control criteria to construct high-coverage global SSH gridded products, the test results spanning three distinct months demonstrated significant performance improvements across all constellations. Following the error correction model optimization, the FY-3G constellation achieved a corrected Mean Absolute Error (MAE) ranging from 1.433 m to 2.158 m, representing a reduction of 40%–70% compared to pre-correction values. Similarly, the corrected MAE for the CYGNSS constellation ranged from 2.178 m to 4.192 m, also reflecting a 40%–70% reduction. For the Tianmu-1 constellation, the corrected MAE was refined to 1.311 m - 1.505 m, with an MAE reduction exceeding 80%. The fusion of the four satellite systems achieved a sea surface coverage of 75.75% within an 8-hour window, with high consistency against validation datasets such as DTU18 validation model and ATL12 MSSH. The findings of this study significantly enhance the SSH retrieval accuracy of commercial constellations not originally designed for altimetry purposes (e.g., Tianmu-1) and provide a novel approach for multi-source GNSS-R fusion-based SSH monitoring. This work holds important theoretical significance and practical value, particularly offering broad application prospects in global ocean monitoring and climate change research.

16 April 2026 in Remote Sensing
Verified
Journal Article
Open access
PDF file
SYShuhu Yang
Shuhu Yang
HCHao Chen
Hao Chen
YZYun Zhang
Yun Zhang
QSQingjing Shi
Qingjing Shi
BPBo Peng
Bo Peng
Zhonghua HongZhonghua Hong
Zhonghua Hong
Snow density plays a crucial role in water resource estimation, runoff forecasting, and early warning of natural disasters such as avalanches and blizzards. This study uses optical satellite multispectral reflectance data to retrieve snow density, providing a novel perspective for snow density retrieval research. Supported by auxiliary data including CanSWE in situ measurements, Sentinel-2 satellite data, and ERA5-Land reanalysis data, this study constructs a hybrid model (Snow_ACMix) that integrates the strengths of the multi-head attention mechanism and convolutional neural networks, realizing direct snow density retrieval from multispectral satellite reflectance data for the first time. This research was primarily conducted in Canada and Alaska. For the Canadian region, the model achieves a mean absolute error (MAE) of 0.034 g/cm3, a root mean square error (RMSE) of 0.051 g/cm3, and a coefficient of determination (R2) of 0.547. For the Alaska region, the model yields an MAE of 0.020 g/cm3, an RMSE of 0.029 g/cm3, and an R2 of 0.803. Feature and module ablation experiments are carried out, and one-shot transfer learning is adopted to perform snow density retrieval in the Alaska region. The spatial transfer prediction results show an MAE of 0.027 g/cm3, an RMSE of 0.038 g/cm3, and an R2 of 0.747, which verify the model’s excellent spatial generalization ability and superior performance in data-scarce regions. The advantages and limitations of the Snow_ACMix model are investigated through comparative validation across different land cover types, regions, time periods, and against ERA5 data. The Snow_ACMix model achieves favorable retrieval performance in mountainous areas, and its practical application capability is verified by snow density retrieval in the Silver Star Mountain region. However, the model still has limitations: it is vulnerable to the effects of wet snow, resulting in large fluctuations in retrieval results in wet snow regions.