Mr. Artyom S. Makovetskii Profile

Artyom S. Makovetskii

at Chelyabinsk State Univ

SPIE Involvement:

Author

Publications (43)

Proceedings Article | 4 October 2023 Poster + Paper

Deep neural network for incongruent point clouds registration

Sergei Voronin, Alexander Vasilyev, Vitaly Kober, Artyom Makovetskii, Aleksei Voronin, Dmitrii Zhernov

Proceedings Volume 12674, 126741L (2023) https://doi.org/10.1117/12.2677782

KEYWORDS: Point clouds, 3D modeling, Clouds, Neural networks, Databases, Education and training, Matrices, Evolutionary algorithms, Network architectures, Singular value decomposition

Read Abstract +

Proceedings Article | 4 October 2023 Poster + Paper

Neural network analog of the ICP algorithm

Sergey Leonov, Alexander Vasilyev, Artyom Makovetskii, Egor Gordionok, Vitaly Kober

Proceedings Volume 12674, 126741H (2023) https://doi.org/10.1117/12.2676383

KEYWORDS: Clouds, Matrices, Neural networks, Evolutionary algorithms, Artificial neural networks, Gallium nitride, Visualization, Point clouds, Analog electronics, Reservoir computing

Read Abstract +

The paper deals with the problem of overlapping two point-data clouds. Traditionally, iterative or variational methods are used to solve such problems. However, these methods are ineffective to solve tasks with a large number of points in the clouds or in the cases of task series with real-time cloud mapping. For those tasks, it is more appropriate to use neural network technique and deep learning methods. The overlap of point-data clouds is understood as finding the displacement vector between them and the rotation matrix of the clouds relative to each other. First of all, point-data clouds are reduced to the zero displacement by means of some transformation. To find the rotation matrix for transformed clouds the authors proposed a simple neural network implementation of the ICP algorithm. This implementation consists of two stages substantially formed by neural networks. At the first stage, a two-layer probabilistic network acts as a metric classifier. The first layer of the probabilistic network is composed of radial-basis elements – Gaussians. The Gaussian activation function makes it possible to identify the output of the first layer with the probability showing the proximity of the points of the superimposed clouds. The second layer of this network is competitive. As a result of the probabilistic network, the points of these two clouds are ranked according to the degree of proximity. The point-data clouds sorted by proximity are sent to the second single-layer neural network. On the second stage, the rotation matrix is calculated using the learning procedure according to the Hebb rule. In the case of small point clouds (less than 10 thousand points), it is more appropriate to use a pseudo-inverse rule (calculations using a pseudoinverse Penrose-Moore matrix) based on the Hebb rule. At the output of the second stage, the rotation matrix is obtained, with which we can easily calculate the displacement vector of the original point clouds. The approbation of the proposed point-data cloud overlap method showed a good match on samples from the ModelNet40 database.

Proceedings Article | 4 October 2023 Poster + Paper

Convolutional auto-encoder to extract local features of 2D images

Vitaly Kober, Sergei Voronin, Artyom Makovetskii, Dmitrii Zhernov, Aleksei Voronin

Proceedings Volume 12674, 126741J (2023) https://doi.org/10.1117/12.2677848

KEYWORDS: Mammography, Neural networks, Feature extraction, Breast cancer, Education and training, Matrices, Image segmentation, Image classification, Classification systems, Imaging systems

Read Abstract +

Proceedings Article | 4 October 2023 Poster + Paper

Multiple point cloud registration and global consistency condition

Artyom Makovetskii, Sergei Voronin, Vitaly Kober, Aleksei Voronin, Tatyana Makovetskaya

Proceedings Volume 12674, 126741I (2023) https://doi.org/10.1117/12.2677107

KEYWORDS: Point clouds, 3D modeling, Clouds, Matrices, Databases, Computer simulations, Reconstruction algorithms, Singular value decomposition, Sensors, Mobile robots

Read Abstract +

Proceedings Article | 3 October 2022 Poster + Paper

Neural network for 3D point clouds alignment

Sergei Voronin, Alexander Vasilyev, Vitaly Kober, Artyom Makovetskii, Aleksei Voronin, Dmitrii Zhernov

Proceedings Volume 12226, 122261H (2022) https://doi.org/10.1117/12.2633565

KEYWORDS: Clouds, Neural networks, Network architectures, Evolutionary algorithms, Transformers, Feature extraction, Computer simulations, 3D image processing, Matrices, Mathematics

Read Abstract +

Proceedings Article | 3 October 2022 Poster + Paper

ICP error functional using point cloud geometry

Artyom Makovetskii, Sergei Voronin, Vitaly Kober, Aleksei Voronin

Proceedings Volume 12226, 122261G (2022) https://doi.org/10.1117/12.2633559

KEYWORDS: Clouds, Computer simulations, Matrices, Reconstruction algorithms, Sensors, Machine vision, Databases, Computer vision technology, Chemical elements, Vision geometry

Read Abstract +

Proceedings Article | 3 October 2022 Poster + Paper

Clustering in coarse registration task and extraction of common parts of point clouds

Sergei Voronin, Artyom Makovetskii, Vitaly Kober, Aleksei Voronin, Mikhail Turov

Proceedings Volume 12226, 122261F (2022) https://doi.org/10.1117/12.2633548

KEYWORDS: Clouds, Matrices, 3D modeling, Data modeling, Reconstruction algorithms, Machine vision, Databases, Computer vision technology, Computer simulations, 3D vision

Read Abstract +

Proceedings Article | 1 August 2021 Poster + Paper

Regularized variational functional use a rough alignment for point clouds registration

Artyom Makovetskii, Sergei Voronin, Vitaly Kober, Aleksei Voronin

Proceedings Volume 11842, 1184221 (2021) https://doi.org/10.1117/12.2594561

KEYWORDS: Clouds, 3D modeling, Computer simulations, Matrices, Data modeling, Reconstruction algorithms, Mathematics, Databases, Computer science

Read Abstract +

Proceedings Article | 1 August 2021 Poster + Paper

ICP algorithm based on stochastic approach

Sergei Voronin, Artyom Makovetskii, Vitaly Kober, Aleksei Voronin

Proceedings Volume 11842, 118421P (2021) https://doi.org/10.1117/12.2594567

KEYWORDS: Clouds, Data modeling, 3D modeling, Stochastic processes, Reconstruction algorithms, Mathematical modeling, Detection and tracking algorithms, Computer simulations, Data processing, Visibility

Read Abstract +

Proceedings Article | 1 August 2021 Poster + Paper

Convolutional neural network for 3D point clouds matching

Sergei Voronin, Artyom Makovetskii, Aleksei Voronin, Dmitrii Zhernov

Proceedings Volume 11842, 1184222 (2021) https://doi.org/10.1117/12.2594573

KEYWORDS: Clouds, Neural networks, Data modeling, 3D modeling, Detection and tracking algorithms, Solid modeling, Databases, Convolutional neural networks, Computer simulations, Robotics

Read Abstract +

Proceedings Article | 21 August 2020 Poster + Paper

An efficient algorithm of total variation regularization in the two-dimensional case

Artyom Makovetskii, Sergei Voronin, Vitaly Kober, Aleksei Voronin

Proceedings Volume 11510, 115102Y (2020) https://doi.org/10.1117/12.2568954

KEYWORDS: Clouds, Image restoration, Optical filters, Computer simulations, Cameras, Matrices, Image denoising, Denoising, Sensors

Read Abstract +

Proceedings Article | 21 August 2020 Paper

Neural network and non-rigid ICP in facial recognition problem

Sergei Voronin, Artyom Makovetskii, Aleksei Voronin, Dmitrii Zhernov

Proceedings Volume 11510, 115102Z (2020) https://doi.org/10.1117/12.2568961

KEYWORDS: Facial recognition systems, 3D modeling, Neural networks, Databases, Convolutional neural networks

Read Abstract +

Proceedings Article | 21 August 2020 Paper

Image dehazing based on microscanning approach

Sergei Voronin, Artyom Makovetskii, Vitaly Kober, Aleksei Voronin, Tatyana Makovetskaya

Proceedings Volume 11510, 115102X (2020) https://doi.org/10.1117/12.2568946

KEYWORDS: Image processing, Image restoration, Imaging systems, Image sensors, Computer simulations

Read Abstract +

Proceedings Article | 6 September 2019 Paper

Non-rigid ICP and 3D models for face recognition

Sergei Voronin, Vitaly Kober, Artyom Makovetskii, Aleksei Voronin

Proceedings Volume 11137, 111372K (2019) https://doi.org/10.1117/12.2529525

KEYWORDS: Facial recognition systems, 3D modeling, Clouds, 3D image processing, 3D scanning, Databases, Computer simulations, Image registration

Read Abstract +

Proceedings Article | 6 September 2019 Paper

Image dehazing using spatially displaced sensors

Sergei Voronin, Vitaly Kober, Artyom Makovetskii, Aleksei Voronin

Proceedings Volume 11137, 111372M (2019) https://doi.org/10.1117/12.2529684

KEYWORDS: Image sensors, Sensors, Atmospheric particles, Image processing, Image restoration, Scattering, Computer simulations, Image analysis

Read Abstract +

Proceedings Article | 6 September 2019 Paper

An efﬁcient 3D mapping framework

Dmitrii Tihonkih, Vitaly Kober, Artyom Makovetskii, Aleksei Voronin

Proceedings Volume 11137, 111372L (2019) https://doi.org/10.1117/12.2529632

KEYWORDS: Clouds, 3D modeling, Reconstruction algorithms, 3D acquisition, Computer simulations, Robotics, Sensors

Read Abstract +

Proceedings Article | 6 September 2019 Paper

Analysis of the convolutional neural network architectures in image classification problems

Sergey Leonov, Alexander Vasilyev, Artyom Makovetskii, Vitaly Kober

Proceedings Volume 11137, 111372E (2019) https://doi.org/10.1117/12.2529232

KEYWORDS: RGB color model, Convolutional neural networks, Image classification, Neural networks, Facial recognition systems, Neurons, Convolution, Databases, Network architectures, Image processing

Read Abstract +