Fast HEVC intra-CU decision partition algorithm with modified LeNet-5 and AlexNet

Imen, Werda; Amna, Maraoui; Belghith, Fatma; Sayadi, Fatma Ezahra; Nouri, Mohammad

doi:10.1007/s11760-022-02139-w

Cited by 19 publications

(11 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Li 15 , et al & Imen 16 , et al propose using CNN-based algorithms as an alternative to the recursive search used in RDO for coding unit partitioning. This strategy maintains compressionefficiency while significantlyreducing the time required to determine if all depths need to be split.…”

Section: Literature Reviewmentioning

confidence: 99%

High Speed Coding Unit Depth Identification Based on Texture Image Information Using SVM

Hari,

Batta

2024

Def. Sc. J.

View full text Add to dashboard Cite

High-Efficiency Video Coding (HEVC/H.265) is a new video coding standard with half the bit rate of its predecessor, Advanced Video Coding (AVC/H.264). AVC/H.264 uses macroblocks, processing units between 4×4 and 16×16 pixels in size. H.265 uses Coding Tree Units (CTUs), a more complicated block structure that lets images be as large as 64×64 pixels. However, changing from macroblocks to coding tree units is essential for H.265 to become more efficient. Using the quadtree structure to divide the Coding Unit (CU) makes it harder for HEVC to find the optimal rate distortion. This paper presents a Support Vector Machine (SVM)-based method for finding the fastest coding unit division in intra-prediction HEVC without compromising compression efficiency. All partitions of CTU are assessed using five characteristics: Standard Deviation (SD), Root Mean Square Error (RMSE), Sub CU Complexity Difference (SCCD), Directional Complexity (DC), and Quantization Parameter (QP) to optimize the intra-prediction of HEVC in all intra-configurations. Simulations have been carried out to estimate the performance of the proposed machine learning-based algorithm using test sequences with different resolutions. Simulation results have shown that combining directional complexity and standard deviation gives a more accurate classification. SVM has been used to separate split-unsplit samples, and the standard rate-distortion optimization technique has been used to separate samples that are hard to separate. The results have shown a reduction of 67.44% in encoding time with a slight increase in bit rate.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

High Speed Coding Unit Depth Identification Based on Texture Image Information Using SVM

Hari,

Batta

2024

Def. Sc. J.

View full text Add to dashboard Cite

show abstract

“…The innovation of convolutional neural networks (CNN), especially the development of the AlexNet [261,262] by Krizhevsky et al [263], validated neural networks as practical universal approximators with layer-wise feature propagation. In CNN, the dense connections of MLPs are replaced with kernel convolutions:…”

Section: Convolutional Neural Network (Cnn)mentioning

confidence: 99%

A Survey of Deep Learning for Alzheimer’s Disease

Zhou

Wang

et al. 2023

MAKE

View full text Add to dashboard Cite

Alzheimer’s and related diseases are significant health issues of this era. The interdisciplinary use of deep learning in this field has shown great promise and gathered considerable interest. This paper surveys deep learning literature related to Alzheimer’s disease, mild cognitive impairment, and related diseases from 2010 to early 2023. We identify the major types of unsupervised, supervised, and semi-supervised methods developed for various tasks in this field, including the most recent developments, such as the application of recurrent neural networks, graph-neural networks, and generative models. We also provide a summary of data sources, data processing, training protocols, and evaluation methods as a guide for future deep learning research into Alzheimer’s disease. Although deep learning has shown promising performance across various studies and tasks, it is limited by interpretation and generalization challenges. The survey also provides a brief insight into these challenges and the possible pathways for future studies.

show abstract

“…AlexNet also has expanded applications to video coding. For example, Imen, et al [61] combined LeNet-5 and AlexNet. Their approach reduces the computational time used to examine block decision candidates.…”

Section: Applicationsmentioning

confidence: 99%

Review of AlexNet for Medical Image Classification

Tang,

Sun,

Wang

et al. 2023

EAI Endorsed Trans e-Learn

View full text Add to dashboard Cite

In recent years, the rapid development of deep learning has led to a wide range of applications in the field of medical image classification. The variants of neural network models with ever-increasing performance share some commonalities: to try to mitigate overfitting, improve generalization, avoid gradient vanishing and exploding, etc. AlexNet first utilizes the dropout technique to mitigate overfitting and the ReLU activation function to avoid gradient vanishing. Therefore, we focus our discussion on AlexNet, which has contributed greatly to the development of CNNs in 2012. After reviewing over 40 papers, including journal papers and conference papers, we give a narrative on the technical details, advantages, and application areas of AlexNet.

show abstract

Fast HEVC intra-CU decision partition algorithm with modified LeNet-5 and AlexNet

Cited by 19 publications

References 14 publications

High Speed Coding Unit Depth Identification Based on Texture Image Information Using SVM

High Speed Coding Unit Depth Identification Based on Texture Image Information Using SVM

A Survey of Deep Learning for Alzheimer’s Disease

Review of AlexNet for Medical Image Classification

Contact Info

Product

Resources

About