Category Ranking
98%
Total Visits
921
Avg Visit Duration
2 minutes
Citations
20
Article Abstract
In this paper, we propose a complex-valued attention feature distillation network (CAFDN) that incorporates a novel lightweight module (CAFDN_Lite) within dual U-Net architectures for phase-only hologram (POH) generation. The proposed network architecture employs simplified upsampling and downsampling layers to enhance computational efficiency, while the CAFDN_Lite module implements hierarchical feature extraction through integrated local attention mechanisms, multi-scale analysis, and channel pruning. This synergistic design enables progressive feature refinement across successive network layers, achieving optimal balance between representational capacity and computational efficiency through systematic feature distillation. The proposed method achieves an average peak signal-to-noise ratio (PSNR) of 32.52 dB and an average structural similarity index (SSIM) of 0.861 within a running time of 36 ms, outperforming conventional approaches. Both numerical reconstructions and optical experiments confirm superior detail reproduction and image quality while reducing computational demands. These advancements highlight the framework's potential for practical holographic display applications, especially in real-time and high-fidelity reconstruction scenarios.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1364/OE.563599 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Incremental Learning for Defect Segmentation With Efficient Transformer Semantic Complement.
IEEE Trans Neural Netw Learn Syst
September 2025
In industrial scenarios, semantic segmentation of surface defects is vital for identifying, localizing, and delineating defects. However, new defect types constantly emerge with product iterations or process updates. Existing defect segmentation models lack incremental learning capabilities, and direct fine-tuning (FT) often leads to catastrophic forgetting.
View Article and Find Full Text PDFGround Reaction Force Estimation via Time-aware Knowledge Distillation.
IEEE Internet Things J
August 2025
Geometric Media Lab, School of Arts, Media and Engineering and School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, AZ 85281 USA.
Human gait analysis with wearable sensors has been widely used in various applications, such as daily life healthcare, rehabilitation, physical therapy, and clinical diagnostics and monitoring. In particular, ground reaction force (GRF) provides critical information about how the body interacts with the ground during locomotion. Although instrumented treadmills have been widely used as the gold standard for measuring GRF during walking, their lack of portability and high cost make them impractical for many applications.
View Article and Find Full Text PDF4D trajectory lightweight prediction algorithm based on knowledge distillation technique.
Front Neurorobot
August 2025
College of Air Traffic Management, Civil Aviation Flight University of China, Chengdu, China.
Introduction: To address the challenges of current 4D trajectory prediction-specifically, limited multi-factor feature extraction and excessive computational cost-this study develops a lightweight prediction framework tailored for real-time air-traffic management.
Methods: We propose a hybrid RCBAM-TCN-LSTM architecture enhanced with a teacher-student knowledge distillation mechanism. The Residual Convolutional Block Attention Module (RCBAM) serves as the teacher network to extract high-dimensional spatial features via residual structures and channel-spatial attention.
SelfReg-UNet: Self-Regularized UNet for Medical Image Segmentation.
Med Image Comput Comput Assist Interv
October 2024
School of Computing and Augmented Intelligence, Arizona State University, AZ, USA.
Since its introduction, UNet has been leading a variety of medical image segmentation tasks. Although numerous follow-up studies have also been dedicated to improving the performance of standard UNet, few have conducted in-depth analyses of the underlying interest pattern of UNet in medical image segmentation. In this paper, we explore the patterns learned in a UNet and observe two important factors that potentially affect its performance: (i) irrelative feature learned caused by asymmetric supervision; (ii) feature redundancy in the feature map.
View Article and Find Full Text PDFKGG: Knowledge-Guided Graph Self-Supervised Learning to Enhance Molecular Property Predictions.
J Chem Inf Model
September 2025
Faculty of Pharmacy, University of Medicine and Pharmacy at Ho Chi Minh City, 41 Dinh Tien Hoang, District 1, Ho Chi Minh City 700000, Vietnam.
Molecular property prediction has become essential in accelerating advancements in drug discovery and materials science. Graph Neural Networks have recently demonstrated remarkable success in molecular representation learning; however, their broader adoption is impeded by two significant challenges: (1) data scarcity and constrained model generalization due to the expensive and time-consuming task of acquiring labeled data and (2) inadequate initial node and edge features that fail to incorporate comprehensive chemical domain knowledge, notably orbital information. To address these limitations, we introduce a Knowledge-Guided Graph (KGG) framework employing self-supervised learning to pretrain models using orbital-level features in order to mitigate reliance on extensive labeled data sets.
View Article and Find Full Text PDF