检索结果-内蒙古大学图书馆

Hardware-Accelerated Event-Graph Neural Networks for Low-Latency Time-Series Classification on SoC FPGA 21st

学校读者我要写书评

暂无评论

Hardware-Accelerated Event-Graph Neural Networks for Low-La...

21st International Symposium on Applied Reconfigurable Computing, ARC 2025

作者： Nakano, Hiroshi Blachut, Krzysztof Jeziorek, Kamil Wzorek, Piotr Dampfhoffer, Manon Mesquida, Thomas Nishi, Hiroaki Kryjak, Tomasz Dalgaty, Thomas Graduate School of Science and Technology Keio University Tokyo Japan Embedded Vision Systems Group Computer Vision Laboratory AGH University of Krakow Krakow Poland CEA-List Université Grenoble Alpes Grenoble France

ISBN: (纸本)9783031879944

As the quantities of data recorded by embedded edge sensors grow, so too does the need for intelligent local processing. Such data often comes in the form of time-series signals, based on which real-time predictions can be made locally using an AI model. However, a hardware-software approach capable of making low-latency predictions with low power consumption is required. In this paper, we present a hardware implementation of an event-graph neural network for time-series classification. We leverage an artificial cochlea model to convert the input time-series signals into a sparse event-data format that allows the event-graph to drastically reduce the number of calculations relative to other AI methods. We implemented the design on a SoC FPGA and applied it to the real-time processing of the Spiking Heidelberg Digits (SHD) dataset to benchmark our approach against competitive solutions. Our method achieves a floating-point accuracy of 92.7% on the SHD dataset for the base model, which is only 2.4% and 2% less than the state-of-the-art models with over 10× and 67× fewer model parameters, respectively. It also outperforms FPGA-based spiking neural network implementations by 19.3% and 4.5%, achieving 92.3% accuracy for the quantised model while using fewer computational resources and reducing latency. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： event-based audio processing FPGA graph convolutional neural networks dynamic audio sensor artificial cochlea

Edge-aware Feature Aggregation Network for Polyp Segmentation

学校读者我要写书评

暂无评论

Machine Intelligence Research 2025年第1期22卷 101-116页

作者： Tao Zhou Yizhe Zhang Geng Chen Yi Zhou Ye Wu Deng-Ping Fan PCA Lab Key Laboratory of Intelligent Perception and Systems for High-dimensional Information of Ministry of EducationSchool of Computer Science and EngineeringNanjing University of Science and TechnologyNanjing210094China School of Computer Science and Engineering Northwestern Polytechnical University(NPU)Xi’an710129China School of Computer Science and Engineering Southeast UniversityNanjing211189China Computer Vision Lab ETH ZürichZürich8092Switzerland

Precise polyp segmentation is vital for the early diagnosis and prevention of colorectal cancer(CRC)in clinical ***,due to scale variation and blurry polyp boundaries,it is still a challenging task to achieve satisfactory segmentation performance with different scales and *** this study,we present a novel edge-aware feature aggregation network(EFA-Net)for polyp segmentation,which can fully make use of cross-level and multi-scale features to enhance the performance of polyp ***,we first present an edge-aware guidance module(EGM)to combine the low-level features with the high-level features to learn an edge-enhanced feature,which is incorporated into each decoder unit using a layer-by-layer ***,a scale-aware convolution module(SCM)is proposed to learn scale-aware features by using dilated convolutions with different ratios,in order to effectively deal with scale ***,a cross-level fusion module(CFM)is proposed to effectively integrate the cross-level features,which can exploit the local and global contextual ***,the outputs of CFMs are adaptively weighted by using the learned edge-aware feature,which are then used to produce multiple side-out segmentation *** results on five widely adopted colonoscopy datasets show that our EFA-Net outperforms state-of-the-art polyp segmentation methods in terms of generalization and *** implementation code and segmentation maps will be publicly at https://***/taozh2017/EFANet.

关键词： Colorectal cancer polyp segmentation edge-aware guidance module scale-aware convolution module cross-level fusion module

Real-Time Multi-Object Tracking using YOLOv8 and SORT on a SoC FPGA

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Danilowicz, Michal Kryjak, Tomasz Embedded Vision Systems Group Computer Vision Laboratory Department of Automatic Control and Robotics AGH University of Krakow Poland

关键词： System-on-chip

Learning to see speckle in the weak laser field through multimode fiber

学校读者我要写书评

暂无评论

Optoelectronics Letters 2025年

作者： JI Yunqi SONG Binbin LI Xueqing LI Yonghui The Engineering Research Center of Learning-Based Intelligent System (Ministry of Education)， the Key Laboratory of Computer Vision and Systems (Ministry of Education)， and the School of Computer Science and Engineering， Tianjin University of Technology

Multimode fibers (MMFs) have great potential for endoscopic imaging due to the high number of modes and a small core diameter. Deep learning based on neural networks has received increasing attention in the field of scattering image reconstruction. However， most studies focus on designing complex network architectures to improve reconstruction， but these network models struggle to reconstruct images in a weak laser field. In the paper， a lightweight generative adversarial network model combined with a histogram specification algorithm is designed to reconstruct speckles in the weak laser field through MMF. Experimental results show that the reconstruction results of our algorithm have better metrics. Moreover， the model demonstrates excellent cross-domain generalization ability with regards to the Fashion-MNIST dataset. It is worth mentioning that we found that the speckles after inactivation still retain the ability to be reconstructed， which enhances the robustness of the model

关键词：

KAN-Face: Efficient Resource Usage and Precision Lip-Sync in Talking Head Generation

学校读者我要写书评

暂无评论

KAN-Face: Efficient Resource Usage and Precision Lip-Sync in...

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Feng, Guanwen Jin, Siyu Qian, Zhihao Li, Yunan Miao, Qiguang School of Computer Science and Technology Xidian University Shaanxi Xi'an710071 China Xi'an Key Laboratory of Big Data and Intelligent Vision Xidian University Shaanxi Xi'an710071 China Key Laboratory of Collaborative Intelligence Systems Ministry of Education Xidian University Xi'an710071 China

ISBN: (纸本)9798350368741

Despite significant progress in NeRF-based talking head generation, problems like poor lip synchronization and inefficient resource usage remain. To solve these, we propose KAN-Face, a lightweight framework. In preprocessing, we introduce a Lip-Sync Enhancement Module that uses Wav2Lip to extract high-resolution audio features and map them to an explicit intermediate representation, ensuring precise lip movement alignment with the speaker's identity. These predicted lip-sync features are combined with fundamental audio-extracted lip features and injected into the rendering module to improve synchronization. For rendering, we introduce FastKAN to map spatial points to color values. As a variant of KAN, FastKAN's sensitivity to 3D scenes and efficient structure enable precise, fast color prediction. Our framework reduces resource consumption while enhancing lip-sync accuracy and facial reconstruction, making it ideal for talking head generation tasks in resource-limited settings. Project: https://***/KAN-Face/. © 2025 IEEE.

关键词： Face Animation KAN Network Low Resource Neural Radiance Fields Talking Head Generation

vision-based automatic fruit counting with UAV

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Szolc, Hubert Wasala, Mateusz Mietla, Remigiusz Iwicki, Kacper Kryjak, Tomasz Embedded Vision Systems Group Computer Vision Laboratory Department of Automatic Control and Robotics AGH University of Krakow al. Mickiewicza 30 Krakow30–059 Poland

The use of unmanned aerial vehicles (UAVs) for smart agriculture is becoming increasingly popular. This is evidenced by recent scientific works, as well as the various competitions organised on this topic. Therefore, in this work we present a system for automatic fruit counting using UAVs. To detect them, our solution uses a vision algorithm that processes streams from an RGB camera and a depth sensor using classical image operations. Our system also allows the planning and execution of flight trajectories, taking into account the minimisation of flight time and distance covered. We tested the proposed solution in simulation and obtained an average score of 87.27/100 points from a total of 500 missions. We also submitted it to the UAV Competition organised as part of the ICUAS 2024 conference, where we achieved an average score of 84.83/100 points, placing 6th in a field of 23 teams and advancing to the finals. © 2025, CC BY.

关键词：

Hardware-Accelerated Event-Graph Neural Networks for Low-Latency Time-Series Classification on SoC FPGA

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Nakano, Hiroshi Blachut, Krzysztof Jeziorek, Kamil Wzorek, Piotr Dampfhoffer, Manon Mesquida, Thomas Nishi, Hiroaki Kryjak, Tomasz Dalgaty, Thomas Graduate School of Science and Technology Keio University Tokyo Japan Embedded Vision Systems Group Computer Vision Laboratory AGH University of Krakow Poland CEA-List Université Grenoble Alpes Grenoble France

关键词： Graph neural networks

CamSAM2: Segment Anything Accurately in Camouflaged Videos

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhou, Yuli Sun, Guolei Li, Yawei Fu, Yuqian Benini, Luca Konukoglu, Ender Computer Vision Laboratory ETH Zurich Switzerland University of Zurich Switzerland Integrated Systems Laboratory ETH Zurich Switzerland INSAIT Sofia University "St. Kliment Ohridski" Bulgaria University of Bologna Italy

Video camouflaged object segmentation (VCOS), aiming at segmenting camouflaged objects that seamlessly blend into their environment, is a fundamental vision task with various real-world applications. With the release of SAM2, video segmentation has witnessed significant progress. However, SAM2’s capability of segmenting camouflaged videos is suboptimal, especially when given simple prompts such as point and box. To address the problem, we propose Camouflaged SAM2 (CamSAM2), which enhances SAM2’s ability to handle camouflaged scenes without modifying SAM2’s parameters. Specifically, we introduce a decamouflaged token to provide the flexibility of feature adjustment for VCOS. To make full use of fine-grained and high-resolution features from the current frame and previous frames, we propose implicit object-aware fusion (IOF) and explicit object-aware fusion (EOF) modules, respectively. Object prototype generation (OPG) is introduced to abstract and memorize object prototypes with informative details using high-quality features from previous frames. Extensive experiments are conducted to validate the effectiveness of our approach. While CamSAM2 only adds negligible learnable parameters to SAM2, it substantially outperforms SAM2 on three VCOS datasets, especially achieving 12.2 mDice gains with click prompt on MoCA-Mask and 19.6 mDice gains with mask prompt on SUN-SEG-Hard, with Hiera-T as the backbone. The code will be available at ***/zhoustan/CamSAM2. Copyright © 2025, The Authors. All rights reserved.

关键词：

KAN-Face: Efficient Resource Usage and Precision Lip-Sync in Talking Head Generation

学校读者我要写书评

暂无评论

KAN-Face: Efficient Resource Usage and Precision Lip-Sync in...

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Guanwen Feng Siyu Jin Zhihao Qian Yunan Li Qiguang Miao School of Computer Science and Technology Xidian University Xi’an China Xi’an Key Laboratory of Big Data and Intelligent Vision Xidian University Xi’an China Key Laboratory of Collaborative Intelligence Systems Ministry of Education Xidian University Xi’an China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Despite significant progress in NeRF-based talking head generation, problems like poor lip synchronization and inefficient resource usage remain. To solve these, we propose KANFace, a lightweight framework. In preprocessing, we introduce a Lip-Sync Enhancement Module that uses Wav2Lip to extract high-resolution audio features and map them to an explicit intermediate representation, ensuring precise lip movement alignment with the speaker’s identity. These predicted lip-sync features are combined with fundamental audio-extracted lip features and injected into the rendering module to improve synchronization. For rendering, we introduce FastKAN to map spatial points to color values. As a variant of KAN, FastKAN’s sensitivity to 3D scenes and efficient structure enable precise, fast color prediction. Our framework reduces resource consumption while enhancing lip-sync accuracy and facial reconstruction, making it ideal for talking head generation tasks in resource-limited settings. Project: https://***/KAN-Face/

关键词： Accuracy Three-dimensional displays Lips Color Signal processing Rendering (computer graphics) Feature extraction Synchronization Speech processing Faces