Welcome to 3D Image Simulation and Recognition Lab of USTB !

Research Gallery

Congratulations on the Recent Paper Acceptances!🎉🎉🎉

  • Cheng Yu, Jiansheng Chen, Yu Wang, Youze Xue, Huimin Ma, Improving Adversarial Robustness Against Universal Patch Attacks Through Feature Norm Suppressing, IEEE Transactions on Neural Networks and Learning Systems, 36(1), pp. 1410-1424 , Jan. 2025 (“By suppressing the feature magnitudes of convolutional neural networks, the image classifier and object detector are made resilient to attacks from adversarial patches in the physical world.”) 中科院一区期刊
  • Ting Tang, Jiansheng Chen, Yiqing Huang, Huimin Ma, Yudong Zhang, Hongwei Yu, Image Paragraph Captioning With Topic Clustering And Topic Shift Prediction, Knowledge-Based Systems,286 (2024): 111401 (“By explicitly constructing topic representations and modeling the probability distribution of topic transitions in human expressions, we achieve image paragraph interpretations that align more closely with human expressive habits.”) 中科院一区期刊
  • Jiawei Li, Hongwei Yu, Jiansheng Chen, Xinlong Ding, Jinlong Wang, Jinyuan Liu, Bochao Zou, Huimin Ma, A^2RNet: Adversarial Attack Resilient Network for Robust Infrared and Visible Image Fusion, The 39th Annual AAAI Conference on Artificial Intelligence (AAAI), Feb. 2025 (“By introducing a Transformer-based defense refinement module, the fusion of visible light images and infrared images becomes more robust against adversarial noise, ensuring that the fusion results perform well in downstream tasks under adversarial conditions.”) CCF-A国际会议
  • Hongwei Yu, Jiansheng Chen, Xinlong Ding, Yudong Zhang, Ting Tang, Huimin Ma, Step Vulnerability Guided Mean Fluctuation Adversarial Attack against Conditional Diffusion Models, The 38th Annual AAAI Conference on Artificial Intelligence (AAAI), Feb. 2024 (“We propose an adversarial attack method targeting diffusion models and reveal through mathematical modeling that different steps of the reverse process in diffusion models exhibit varying degrees of vulnerability to adversarial attacks.”) CCF-A国际会议
  • Xinlong Ding, Jiansheng Chen, Hongwei Yu, Yu Shang, Yining Qin, Huimin Ma, Transferable Adversarial Attacks for Object Detection using Object-Aware Significant Feature Distortion, The 38th Annual AAAI Conference on Artificial Intelligence (AAAI), Feb. 2024 (“We reveal the presence of significant features in the object detector that are highly correlated with the object areas, and distorting these features can enable effective transfer attacks among object detectors.”) CCF-A国际会议
  • Jiehui Wu, Jiansheng Chen, Qifeng Luo, Siqi Liu, Youze Xue, Huimin Ma, Upper-body Hierarchical Graph for Skeleton Based Emotion Recognition in Assistive Driving, European Conference on Computer Vision (ECCV), Sept. 2024. (“By applying HD-GCN proposed by J. Lee et. al. to upper human body only, we achieved high-performance human action emotion recognition under driving conditions.”) CCF-B国际会议
  • Youze Xue, Jiansheng Chen, Hongbing Ma, Huimin Ma, Refining 3d Human Mesh Via Model-Free Offsets Estimation, IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), April 2024 （”We proposed a framework that integrates model-agnostic and model-free pose estimation methods, achieving more accurate human keypoint pose estimation.”）CCF-B国际会议
  • Jiawei Li, Jiansheng Chen, Jinyuan Liu, Huimin Ma, Learning a Graph Neural Network with Cross Modality Interaction for Image Fusion, ACM Multimedia Conference 2023, Oct. 2023 (“We incorporated cross-modal interaction units into the graph neural network, achieving superior detail representation in the fusion results of visible light and infrared images.”) CCF-A国际会议
  • Yu H, Gong W, Chen J, et al. HomuGAN: A 3D-aware GAN with the Method of Cylindrical Spatial-Constrained Sampling[J]. IEEE Transactions on Image Processing, 2024.
  • Yu H, Gong W, Chen J, et al. GET3DGS: Generate 3D Gaussians Based on Points Deformation Fields[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2024.
  • Song K, Tan Z, Zou B, et al. Unveiling the Dynamics of Information Interplay in Supervised Learning[C]//Forty-first International Conference on Machine Learning. ICML, 2024.
  • Ding X, Chen J, Yu H, et al. Enhancing Adversarial Transferability in Object Detection with Bidirectional Feature Distortion[C]// ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2024.
  • Chen S, Wang R, Li X, et al. PLS: Unsupervised Domain Adaptation for 3D Object Detection via Pseudo-Label Sizes[C]// ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2024.
  • Ding X, Chen J, Yu H, et al. Transferable Adversarial Attacks for Object Detection using Object-Aware Significant Feature Distortion[C]// Proceedings of the AAAI Conference on Artificial Intelligence. 2024.
  • Chen J, Qin Y, Lin P, et al. Center of Pressure Estimation by Analyzing Walking Videos[C]// ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2024.
  • Yu H, Chen J, Ding X, et al. Step Vulnerability Guided Mean Fluctuation Adversarial Attack against Conditional Diffusion Models[C]// Proceedings of the AAAI Conference on Artificial Intelligence. 2024.

Congratulations on the Recent Competitions and workshops!💯💯💯

  • Li X, et al. 2nd place winner of the 2024 Ego4D-Ego-Exo4D Challenge @ CVPR 2024 in the Looking at Me track. 
  • Guo Z, et al. Third place of track 2 of the 3rd Vision-based Remote Physiological Signal Sensing (RePSS) Challenge & Workshop.