AIMC Topic: Data Compression

Clear Filters Showing 51 to 60 of 147 articles

Low-Complexity Adaptive Sampling of Block Compressed Sensing Based on Distortion Minimization.

Sensors (Basel, Switzerland) Jun 25, 2022
Block compressed sensing (BCS) is suitable for image sampling and compression in resource-constrained applications. Adaptive sampling methods can effectively improve the rate-distortion performance of BCS. However, adaptive sampling methods bring hig...
Neural Networks, Computer Image Processing, Computer-Assisted Data Compression
View on PubMed DOI

Towards Convolutional Neural Network Acceleration and Compression Based on -Means.

Sensors (Basel, Switzerland) Jun 6, 2022
Convolutional Neural Networks (CNNs) are popular models that are widely used in image classification, target recognition, and other fields. Model compression is a common step in transplanting neural networks into embedded devices, and it is often use...
Neural Networks, Computer Algorithms Acceleration Data Compression Physical Phenomena
View on PubMed DOI

MobilePrune: Neural Network Compression via Sparse Group Lasso on the Mobile System.

Sensors (Basel, Switzerland) May 27, 2022
It is hard to directly deploy deep learning models on today's smartphones due to the substantial computational costs introduced by millions of parameters. To compress the model, we develop an ℓ0-based sparse group lasso model called MobilePrune which...
Neural Networks, Computer Algorithms Data Compression Physical Phenomena
View on PubMed DOI

LAP: Latency-aware automated pruning with dynamic-based filter selection.

Neural networks : the official journal of the International Neural Network Society May 10, 2022
Model pruning is widely used to compress and accelerate convolutional neural networks (CNNs). Conventional pruning techniques only focus on how to remove more parameters while ensuring model accuracy. This work not only covers the optimization of mod...
Neural Networks, Computer Algorithms Automation Data Compression
View on PubMed DOI

Progressive compressive sensing of large images with multiscale deep learning reconstruction.

Scientific reports May 4, 2022
Compressive sensing (CS) is a sub-Nyquist sampling framework that has been employed to improve the performance of numerous imaging applications during the last 15 years. Yet, its application for large and high-resolution imaging remains challenging i...
Humans Algorithms Deep Learning User-Computer Interface Imaging, Three-Dimensional Data Compression Physical Phenomena
View on PubMed DOI

StructADMM: Achieving Ultrahigh Efficiency in Structured Pruning for DNNs.

IEEE transactions on neural networks and learning systems May 2, 2022
Weight pruning methods of deep neural networks (DNNs) have been demonstrated to achieve a good model pruning rate without loss of accuracy, thereby alleviating the significant computation/storage requirements of large-scale DNNs. Structured weight pr...
Neural Networks, Computer Data Compression
View on PubMed DOI

Communication-efficient federated learning via knowledge distillation.

Nature communications Apr 19, 2022
Federated learning is a privacy-preserving machine learning technique to train intelligent models from decentralized data, which enables exploiting private data by communicating local model updates in each iteration of model learning rather than the ...
Humans Machine Learning Communication Privacy Data Compression
View on PubMed DOI

SensiMix: Sensitivity-Aware 8-bit index & 1-bit value mixed precision quantization for BERT compression.

PloS one Apr 18, 2022
Given a pre-trained BERT, how can we compress it to a fast and lightweight one while maintaining its accuracy? Pre-training language model, such as BERT, is effective for improving the performance of natural language processing (NLP) tasks. However, ...
Language Natural Language Processing Data Compression
View on PubMed DOI

Deep-learning-based projection-domain breast thickness estimation for shape-prior iterative image reconstruction in digital breast tomosynthesis.

Medical physics Mar 30, 2022
BACKGROUND: Digital breast tomosynthesis (DBT) is a technique that can overcome the shortcomings of conventional X-ray mammography and can be effective for the early screening of breast cancer. The compression of the breast is essential during the DB...
Humans Algorithms Female Breast Neoplasms Deep Learning Image Processing, Computer-Assisted Mammography Breast Phantoms, Imaging Data Compression
View on PubMed DOI

Skeleton-Based Spatio-Temporal U-Network for 3D Human Pose Estimation in Video.

Sensors (Basel, Switzerland) Mar 28, 2022
Despite the great progress in 3D pose estimation from videos, there is still a lack of effective means to extract spatio-temporal features of different granularity from complex dynamic skeleton sequences. To tackle this problem, we propose a novel, s...
Humans Neural Networks, Computer Semantics Pressure Data Compression Skeleton
View on PubMed DOI
Popular Topics
Recent Journals