检索结果-内蒙古大学图书馆

Modeling Spoof Noise by De-spoofing Diffusion and its Application in Face Anti-spoofing

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Bin Zhu, Xiangyu Zhang, Xiaoyu Lei, Zhen Institute of Information Engineering Chinese Academy of Sciences China School of Cyber Security University of Chinese Academy of Sciences China State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science & Innovation Chinese Academy of Sciences China

Face anti-spoofing is crucial for ensuring the security and reliability of face recognition systems. Several existing face anti-spoofing methods utilize GAN-like networks to detect presentation attacks by estimating the noise pattern of a spoof image and recovering the corresponding genuine image. But GAN’s limited face appearance space results in the denoised faces cannot cover the full data distribution of genuine faces, thereby undermining the generalization performance of such methods. In this work, we present a pioneering attempt to employ diffusion models to denoise a spoof image and restore the genuine image. The difference between these two images is considered as the spoof noise, which can serve as a discriminative cue for face anti-spoofing. We evaluate our proposed method on several intra-testing and inter-testing protocols, where the experimental results showcase the effectiveness of our method in achieving competitive performance in terms of both accuracy and generalization. Copyright © 2024, The Authors. All rights reserved.

关键词： Face recognition

Discriminative Clue Alignment Network for Both Image- and Video-Based Person Re-Identification 32

学校读者我要写书评

暂无评论

Discriminative Clue Alignment Network for Both Image- and Vi...

32nd British Machine Vision Conference, BMVC 2021

作者： Hu, Panwen Zhou, Xinyu Huang, Rui The Chinese University of HongKong Guangdong Shenzhen China Shenzhen Institute of Artificial Intelligence and Robotics for Society Guangdong China

The body misalignment problem resulted from various factors, e.g., scale and pose variation, occlusion, etc., has always been a great challenge in person re-identification. It is intuitive to model individual body parts and match them through pose estimation or human parsing, which, however, requires additional annotations and training, and may fail with occlusion. Some recent studies employ self-attention mechanisms to discover the Discriminative Clues (DCs) on the body. But, unlike the body parts, the DCs are not naturally aligned properly. To this end, we propose a Discriminative Clue Alignment Network (DCANet), along with a discriminant constraint, to automatically identify various DCs and then align them into a fixed pattern, without explicitly analyzing body parts. Moreover, such an alignment scheme makes the temporal aggregation of features from video frames extremely simple, because the DCs are effectively aligned across frames. Therefore our method can be easily applied to video-based person re-identification as well. Experiments on several popular public benchmarks show that DCANet can achieve state-of-the-art performance on both image- and video-based re-identification tasks. © 2021. The copyright of this document resides with its authors.

关键词： Computer vision

Monocular Lane Detection Based on Deep Learning: A Survey

学校读者我要写书评

暂无评论

arXiv 2024年

作者： He, Xin Guo, Haiyun Zhu, Kuan Zhu, Bingke Zhao, Xu Fang, Jianwu Wang, Jinqiao School of Artificial Intelligence University of Chinese Academy of Sciences China Foundation Model Research Center Institute of Automation Chinese Academy of Sciences China Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University China Peng Cheng Laboratory China Wuhan AI Research China

Lane detection plays an important role in autonomous driving perception systems. As deep learning algorithms gain popularity, monocular lane detection methods based on them have demonstrated superior performance and emerged as a key research direction in autonomous driving perception. The core designs of these algorithmic frameworks can be summarized as follows: (1) Task paradigm, focusing on lane instance-level discrimination;(2) Lane modeling, representing lanes as a set of learnable parameters in the neural network;(3) Global context supplementation, enhancing inference on the obscure lanes;(4) Perspective effect elimination, providing accurate 3D lanes for downstream applications. From these perspectives, this paper presents a comprehensive overview of existing methods, encompassing both the increasingly mature 2D lane detection approaches and the developing 3D lane detection works. Besides, this paper compares the performance of mainstream methods on different benchmarks and investigates their inference speed under a unified setting for fair comparison. Moreover, we present some extended works on lane detection, including multitask perception, video lane detection, online high-definition map construction, and lane topology reasoning, to offer readers a comprehensive roadmap for the evolution of lane detection. Finally, we point out some potential future research directions in this field. We exhaustively collect the papers and codes of existing works at https://***/Core9724/Awesome-Lane-Detection and will keep tracing the research. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep learning

Functional Connectivity Analysis Revealed Abnormal Brain Network Pattern in Parkinson's Disease with Freezing of Gait

学校读者我要写书评

暂无评论

Functional Connectivity Analysis Revealed Abnormal Brain Net...

第42届中国控制会议

作者： Zhilin Shu Jinrui Liu Ting Song Jianda Han Ningbo Yu College of Artificial Intelligence Nankai University Tianjin Key Laboratory of Intelligent Robotics Nankai University Institute of Intelligence Technology and Robotic Systems Shenzhen Research Institute of Nankai University

Patients with Parkinson's disease(PD) commonly have deficits in lower-limb *** of gait(FOG),a type of gait disturbance of PD,is a complex motor symptom exhibited as inability to generate an effective *** mechanism of lower-limb dysfunction of PD,especially FOG,remains to be *** investigate the neural mechanisms related to FOG in PD,we evaluated 39 subjects(13 PD patients with FOG,13 PD patients without FOG,and 13 healthy subjects) based on electroencephalogram(EEG) signals while they performed the pedaling *** conducted functional connectivity analysis using Magnitude Squared Coherence(MSC) and calculated the metrics to investigate the characteristics of the brain network among the three *** results indicated that PD patients were associated with reduced functional integration and segregation of brain network,subserving both motor and cognitive *** reduced functional integration and segregation of cognitive processes were correlated with the severity of *** study revealed the differences in brain networks between PD patients and healthy controls and provided insights into the neural mechanism of FOG.

关键词：

Autonomous Non-communicative and Vision-based Control Strategy for AGV to Track UAV

学校读者我要写书评

暂无评论

Autonomous Non-communicative and Vision-based Control Strate...

International Conference on Control, Automation and Systems ( ICCAS)

作者： Ashok Kumar Sivarathri Amit Shukla School of Mechanical and Materials Engineering Indian Institute of Technology Mandi Himachal Pradesh India Centre for Artificial Intelligence and Robotics Indian Institute of Technology Mandi Himachal Pradesh India

ISBN: (数字)9788993215380

ISBN: (纸本)9798331517939

UAV-AGV (Unmanned Aerial Vehicle-Autonomous Ground Vehicle) is a suitable system for automation of industrial tasks such as inspection and mapping. Vision-based UAV-AGV system (camera attached to UAV or AGV or both) is prominent in GPS inaccessible zones. Vision-based tracking motion between both agents is an important aspect during execution of outdoor autonomous missions. It can be either UAV tracking AGV or vice-versa. Former one (UAV tracking the motion of AGV) is well solved. Latter one is less focused as per the literature. But it also required for effective collaboration of the two-agents. For UAV leader and AGV follower roles, AGV needs to continuously track UAV for supporting (payload and tethered charging) longer outdoor missions of UAV. However, existing work in the literature is not suitable for the outdoor missions. Hence, this work presents a non-communicative and vision-based tracking strategy for AGV to track the motion of UAV. AGV contains an on-board sky-facing camera to localize UAV. Deep learning-based technique is implemented for robust localization of UAV using on-board camera of AGV. Sliding mode control technique is implemented to develop tracking controller for AGV. UAV is manually operated with random velocity commands and AGV autonomously tracks the projection of center of UAV on the ground plane. Tracking experiments are performed using quadcopter UAV and non-holonomic AGV mobile platforms. Learning based detection of UAV and tracking controller of AGV are showing satisfactory performance from the experimental results. This shows feasibility to execute outdoor missions without requirement of GPS and also communication between the agents.

关键词： Automation Tracking Collaboration Autonomous aerial vehicles Cameras Real-time systems Sliding mode control Global Positioning System Drones Quadrotors

DESIRE: DYNAMIC KNOWLEDGE CONSOLIDATION FOR REHEARSAL-FREE CONTINUAL LEARNING

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Guo, Haiyang Zhu, Fei Zeng, Fanhu Liu, Bing Zhang, Xu-Yao MAIS Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China Centre for Artificial Intelligence and Robotics HKISI-CAS Hong Kong Department of Computer Science University of Illinois Chicago United States

Continual learning aims to equip models with the ability to retain previously learned knowledge like a human. Recent work incorporating Parameter-Efficient Fine-Tuning has revitalized the field by introducing lightweight extension modules. However, existing methods usually overlook the issue of information leakage caused by the fact that the experiment data have been used in pre-trained models. Once these duplicate data are removed in the pre-training phase, their performance can be severely affected. In this paper, we propose a new LoRA-based rehearsal-free method named DESIRE. Our method avoids imposing additional constraints during training to mitigate catastrophic forgetting, thereby maximizing the learning of new classes. To integrate knowledge from old and new tasks, we propose two efficient post-processing modules. On the one hand, we retain only two sets of LoRA parameters for merging and propose dynamic representation consolidation to calibrate the merged feature representation. On the other hand, we propose decision boundary refinement to address classifier bias when training solely on new class data. Extensive experiments demonstrate that our method achieves state-of-the-art performance on multiple datasets and strikes an effective balance between stability and plasticity. Our code will be publicly available. Copyright © 2024, The Authors. All rights reserved.

关键词： Contrastive Learning

Learning Realistic and Reasonable Grasps for Anthropomorphic Hand in Cluttered Scenes

学校读者我要写书评

暂无评论

Learning Realistic and Reasonable Grasps for Anthropomorphic...

IEEE International Conference on robotics and Automation (ICRA)

作者： Haonan Duan Yiming Li Daheng Li Wei Wei Yayu Huang Peng Wang Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China Centre for Artificial Intelligence and Robotics Hong Kong Institutue of Science and Innovation Chinese Academy of Sciences Hong Kong China

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

Grasping is one of the most fundamental skills for humans to interact with objects. However, it remains a challenging problem for anthropomorphic hands, due to the lack of object affordance understanding and high-dimensional grasp planning. In this work, we propose an anthropomorphic hand grasping framework to learn realistic and reasonable grasps in cluttered scenes, which tackles the problem in three items: 1) graspable point segmentation; 2) hand grasp generation and 3) grasp optimization. Specifically, our method generates high-quality hand grasps efficiently without complete object models by learning graspable points, associated grasp configurations from observed point cloud in a parallel manner and optimizing predicted grasps based on hand-object contacts. Simulation experiments show that our model generates physical plausible grasps for the anthropomorphic hand effectively with over 70% success rate. Real-world experiments demonstrate that the model trained in simulation performs satisfactorily in real-world scenarios for unseen objects.

关键词： Point cloud compression Affordances Grasping Predictive models Planning Optimization Robots

Navigating Open Set Scenarios for Skeleton-based Action Recognition

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Peng, Kunyu Yin, Cheng Zheng, Junwei Liu, Ruiping Schneider, David Zhang, Jiaming Yang, Kailun Sarfraz, M. Saquib Stiefelhagen, Rainer Roitberg, Alina Insitute for Anthropomatics and Robotics Karlsruhe Institute of Technology Germany School of Robotics Hunan University China Mercedes-Benz Tech Innovation Germany Institute for Artificial Intelligence University of Stuttgart Germany

In real-world scenarios, human actions often fall outside the distribution of training data, making it crucial for models to recognize known actions and reject unknown ones. However, using pure skeleton data in such open-set conditions poses challenges due to the lack of visual background cues and the distinct sparse structure of body pose sequences. In this paper, we tackle the unexplored Open-Set Skeleton-based Action Recognition (OS-SAR) task and formalize the benchmark on three skeleton-based datasets. We assess the performance of seven established open-set approaches on our task and identify their limits and critical generalization issues when dealing with skeleton information. To address these challenges, we propose a distance-based cross-modality ensemble method that leverages the cross-modal alignment of skeleton joints, bones, and velocities to achieve superior open-set recognition performance. We refer to the key idea as CrossMax - an approach that utilizes a novel cross-modality mean max discrepancy suppression mechanism to align latent spaces during training and a cross-modality distance-based logits refinement method during testing. CrossMax outperforms existing approaches and consistently yields state-of-the-art results across all datasets and backbones. The benchmark, code, and models will be released at https://***/KPeng9510/OS-SAR. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

Leveraging Anchor-based LiDAR 3D Object Detection via Point Assisted Sample Selection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Shitao Zhang, Haolin Zheng, Nanning The National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Shaanxi Xi’an China

3D object detection based on LiDAR point cloud and prior anchor boxes is a critical technology for autonomous driving environment perception and understanding. Nevertheless, an overlooked practical issue in existing methods is the ambiguity in training sample allocation based on box Intersection over Union (IoUbox). This problem impedes further enhancements in the performance of anchor-based LiDAR 3D object detectors. To tackle this challenge, this paper introduces a new training sample selection method that utilizes point cloud distribution for anchor sample quality measurement, named Point Assisted Sample Selection (PASS). This method has undergone rigorous evaluation on two widely utilized datasets. Experimental results demonstrate that the application of PASS elevates the average precision of anchor-based LiDAR 3D object detectors to a novel state-of-the-art, thereby proving the effectiveness of the proposed approach. The codes will be made available at https://***/XJTU-Haolin/Point_Assisted_Sample_Selection. Copyright © 2024, The Authors. All rights reserved.

关键词： Object detection