Neural Video Compression Based on RBM Scene Change Detection Algorithm

Knop, Michał; Kapuściński, Tomasz; Mleczko, Wojciech K.; Angryk, Rafal A.

doi:10.1007/978-3-319-39384-1_58

Cited by 5 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lower-dimensional representations can improve performance on many tasks, such as image compression, reconstruction and clustering (Liu et al, 2019). Color image compression occurs in many information retrieval problems (Dewi et al, 2021;Khanna et al, 2019;Knop et al, 2016). In such tasks, org/10.15405/epct.23021.36 Corresponding Author: V. V. Matskevich Selection and peer-review under responsibility of the Organizing Committee of the conference eISSN: 2672-8834 the restricted Boltzmann machine (RBM) is often used, which allows you to extract informative features that are used later for image classification.…”

Section: The Problem Of Color Images Compression Can Be Formally Writ...mentioning

confidence: 99%

An Efficient Training Algorithm of Restricted Boltzmann Machines

Matskevich¹

2023

European Proceedings of Computers and Technology

View full text Add to dashboard Cite

The paper deals with an actual applied problem related to the artificial neural networks training. An approach to the solution based on the idea of random search is proposed. An original training algorithm that implements Boltzmann annealing has been developed and its convergence in probability to the global optimum has been proved. It is also shown that the proposed algorithm can be easily modified to train any artificial neural network. Thus, it has a good prospect for solving applied problems using neural network technologies in general. Experimental studies have been carried out, in which, using the example of compressing color raster images problem, the proposed algorithm was compared with the known adaptive moment algorithm -one of the best gradient methods for training neural networks. Image compression was performed using an ensemble of n Gauss-Bernoulli restricted Boltzmann machines. The use of an ensemble of n machines in combination with a specially developed parallelization procedure made it possible to reduce the computational complexity of the training process and increase the speed of the proposed algorithm. As a result of experiments, it was shown that the proposed approach is not inferior to gradient methods in terms of speed. Moreover, the developed training algorithm turned out to be more than twice as effective as the adaptive moment algorithm in terms of the quality of the solution obtained.

show abstract

Section: The Problem Of Color Images Compression Can Be Formally Writ...mentioning

confidence: 99%

An Efficient Training Algorithm of Restricted Boltzmann Machines

Matskevich¹

2023

European Proceedings of Computers and Technology

View full text Add to dashboard Cite

show abstract

“…Drozda et al presented proposition of different orderings for visual sequences alignment implemented as algorithms for image classification [3]. Knop et al discussed improvements in application of neural methods into video compression based on dedicated scene change detection algorithm [4], while Capizzi et al proposed novel attempt to process images of oranges to be classified by proposed neural network architecture [5], Stateczny et al discussed application of intelligent methods for image processing in batymetric systems [6]. Intelligent methods are also widely adapted into detection processes: Starzyk developed visual saccades for object recognition [7], Pabiasz et al proposed three-dimensional facial landmarks recognition [8] and novel approach to 3D face images processing [9].…”

Section: A Related Workmentioning

confidence: 99%

Toward adaptive heuristic video frames capturing and correction in real-time

Woźniak

Połap

Capizzi

et al. 2016

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

Abstract-Multimedia devices are widely used in professional applications as well as personal purposes. The use of computer vision systems enables detection and extraction of important features exposed in images. However constantly increasing demand for this type of video with high quality requires simple however reliable methods. The objective of presented research is to investigate applicability of heuristic method for real-time video frames capturing and correction.

show abstract

“…One of the first approaches of this trend were neural networks, which are mainly used as classifiers, i.e. classification of signatures [1], surface structures [2] and multimedia applications [3]. Modern methods of mathematics and artificial intelligence are increasingly finding application in medicine and other aspects.…”

Section: Introductionmentioning

confidence: 99%

Neuro-heuristic voice recognition

Połap

2016

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

Abstract-Protection of private data, signing electronic documents are selective use of identity verification. In this work, the problem of voice verification has been discussed. The method of verifying the voice based on the methods of artificial intelligence was presented. Numerous tests were performed to demonstrate the effectiveness of the presented solution -research results are shown and discussed in terms of advantages and disadvantages.

show abstract

Neural Video Compression Based on RBM Scene Change Detection Algorithm

Cited by 5 publications

References 26 publications

An Efficient Training Algorithm of Restricted Boltzmann Machines

An Efficient Training Algorithm of Restricted Boltzmann Machines

Toward adaptive heuristic video frames capturing and correction in real-time

Neuro-heuristic voice recognition

Contact Info

Product

Resources

About