Mikhail G. Mozerov
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Mikhail G. Mozerov has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Joost van de Weijer
Universitat Autónoma de Barcelona
5 shared publications - Jordi Gonzàlez
Computer Vision Center, Universitat Autònoma de Barcelona, 08193 Barcelona, Spain
4 shared publications - Vitaly Kober
Department of Computer Science, CICESE, Ensenada, Mexico
4 shared publications - Fei Yang
Nankai University
3 shared publications - Chao Zhai
School of Information Science and Engineering, Shandong University, Qingdao, China
2 shared publications Fei Yang
School of Automation Northwestern Polytechnical University Xi'an China
2 shared publications
Latest Publications
- 2 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 11 (6) , 7568-7583
MSSA: Multispectral Semantic Alignment for Cross-Modality Infrared-RGB Person Reidentification
The widespread deployment of dual-camera systems has laid a solid foundation for practical applications of infrared (IR)-RGB cross-modality person reidentification (ReID). However, the inherent modality differences between RGB and IR images cause significant intra-class variances in the feature space for individuals of the same identity. Current methods typically employ various network architectures for the image style transfer or extracting modality-invariant features, yet they overlook the information extraction from the most fundamental spectral semantic features. Based on the existing approaches, we propose a multi-spectral semantic alignment (MSSA) architecture aimed at aligning fine-grained spectral semantic features across both intra-modality and inter-modality perspectives. Through modality center semantic alignment (MCSA) learning, we comprehensively mitigate differences in identity features of different modalities. Moreover, to attenuate the discriminative information unique to a single modality, we introduce the modality reliability intensification (MRI) loss to enhance the reliability of identity information. Finally, to tackle the challenge that inter-modality intra-class disparities surpass inter-modality inter-class differences, we leverage the dynamic discriminative center (DDC) loss to further bolster the discriminability of reliable information. Through an extensive experiments conducted on SYSU-MM01, RegDB, and LLCM datasets, we demonstrate the substantial advantages of the proposed MSSA over other state-of-the-art methods.
- 2 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 11 (6) , 7568-7583
MSSA: Multispectral Semantic Alignment for Cross-Modality Infrared-RGB Person Reidentification
The widespread deployment of dual-camera systems has laid a solid foundation for practical applications of infrared (IR)-RGB cross-modality person reidentification (ReID). However, the inherent modality differences between RGB and IR images cause significant intra-class variances in the feature space for individuals of the same identity. Current methods typically employ various network architectures for the image style transfer or extracting modality-invariant features, yet they overlook the information extraction from the most fundamental spectral semantic features. Based on the existing approaches, we propose a multi-spectral semantic alignment (MSSA) architecture aimed at aligning fine-grained spectral semantic features across both intra-modality and inter-modality perspectives. Through modality center semantic alignment (MCSA) learning, we comprehensively mitigate differences in identity features of different modalities. Moreover, to attenuate the discriminative information unique to a single modality, we introduce the modality reliability intensification (MRI) loss to enhance the reliability of identity information. Finally, to tackle the challenge that inter-modality intra-class disparities surpass inter-modality inter-class differences, we leverage the dynamic discriminative center (DDC) loss to further bolster the discriminability of reliable information. Through an extensive experiments conducted on SYSU-MM01, RegDB, and LLCM datasets, we demonstrate the substantial advantages of the proposed MSSA over other state-of-the-art methods.
- 9 January 2024
- Published by Elsevier in ISPRS Journal of Photogrammetry and Remote Sensing
- Vol. 208, 1-13
A continuous digital elevation representation model for DEM super-resolution
- 9 January 2024
- Published by Elsevier in ISPRS Journal of Photogrammetry and Remote Sensing
- Vol. 208, 1-13
A continuous digital elevation representation model for DEM super-resolution
- 1 December 2023
- Published by Pleiades Publishing Ltd in Journal of Communications Technology and Electronics
- Vol. 68 (12) , 1492-1498
Comparative Analysis of Convolutional Neural Networks for Classification of Breast Abnormalities
—Computer-aided detection (CAD) systems are used by radiologists as a second interpreter for breast cancer detection in digital mammography. However, for every true-positive cancer detected by the CAD system, a large number of false predictions must be reviewed by an expert to avoid an unnecessary biopsy. The traditional approach to creating such systems is to select and compute the features of objects of interest from the source data, followed by the selection of a model for their classification using machine learning. Machine learning and, especially, deep learning are being used to analyze mammograms. Most of the models proposed so far are trained on a small amount of data and do not have high reliability. This paper compares several deep learning models for benign–malign mammography classification on digital mammograms. The preprocessing step is designed to remove noise and extract features using local phase information of the image. Deep learning is then used to classify the digital mammography. The experimental results are presented using several databases and estimated using several quality criteria.
- 1 December 2023
- Published by Pleiades Publishing Ltd in Journal of Communications Technology and Electronics
- Vol. 68 (12) , 1492-1498
Comparative Analysis of Convolutional Neural Networks for Classification of Breast Abnormalities
—Computer-aided detection (CAD) systems are used by radiologists as a second interpreter for breast cancer detection in digital mammography. However, for every true-positive cancer detected by the CAD system, a large number of false predictions must be reviewed by an expert to avoid an unnecessary biopsy. The traditional approach to creating such systems is to select and compute the features of objects of interest from the source data, followed by the selection of a model for their classification using machine learning. Machine learning and, especially, deep learning are being used to analyze mammograms. Most of the models proposed so far are trained on a small amount of data and do not have high reliability. This paper compares several deep learning models for benign–malign mammography classification on digital mammograms. The preprocessing step is designed to remove noise and extract features using local phase information of the image. Deep learning is then used to classify the digital mammography. The experimental results are presented using several databases and estimated using several quality criteria.
- 24 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (24) , 31399-31413
An Unsupervised Domain Adaption Approach for Cross-Modality RGB-Infrared Person Re-Identification
Dual-camera systems commonly employed in surveillance serve as the foundation for RGB-infrared (IR) cross-modality person re-identification (ReID). However, significant modality differences give rise to inferior performance compared to single-modality scenarios. Furthermore, most existing researches in this area rely on supervised training with meticulously labeled datasets. Labeling RGB-IR image pairs is more complex than labeling conventional image data, and deploying pre-trained models on unlabeled datasets can lead to catastrophic performance degradation. In contrast to previous solutions that focus solely on cross-modality or domain adaptation issues, this paper presents an end-to-end unsupervised domain adaptation (UDA) framework for the cross-modality person ReID, which can simultaneously addresses both of these challenges. This model employs source domain classes, target domain clusters, and unclustered instance samples for the training, maximizing the comprehensive use of the dataset. Moreover, it addresses the problem of mismatched clustering labels between the two modalities in the target domain by incorporating a label matching module that reassigns reliable clusters with labels, ensuring correspondence between different modality labels. We construct the loss function by incorporating distinctiveness loss and multiplicity loss, both of which are determined by the similarity of neighboring features in the predicted feature space and the difference between distant features. This approach enables efficient feature clustering and cluster class assignment to occur concurrently. Eight UDA cross-modality person ReID experiments are conducted on three real datasets and six synthetic datasets. Experimental results unequivocally demonstrate that the proposed model outperforms the existing state-of-the-art algorithms to a significant degree. Notably, in RegDB → RegDB_light, the Rank-1 accuracy exhibits a remarkable improvement of 8.24%.
- 24 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 23 (24) , 31399-31413
An Unsupervised Domain Adaption Approach for Cross-Modality RGB-Infrared Person Re-Identification
Dual-camera systems commonly employed in surveillance serve as the foundation for RGB-infrared (IR) cross-modality person re-identification (ReID). However, significant modality differences give rise to inferior performance compared to single-modality scenarios. Furthermore, most existing researches in this area rely on supervised training with meticulously labeled datasets. Labeling RGB-IR image pairs is more complex than labeling conventional image data, and deploying pre-trained models on unlabeled datasets can lead to catastrophic performance degradation. In contrast to previous solutions that focus solely on cross-modality or domain adaptation issues, this paper presents an end-to-end unsupervised domain adaptation (UDA) framework for the cross-modality person ReID, which can simultaneously addresses both of these challenges. This model employs source domain classes, target domain clusters, and unclustered instance samples for the training, maximizing the comprehensive use of the dataset. Moreover, it addresses the problem of mismatched clustering labels between the two modalities in the target domain by incorporating a label matching module that reassigns reliable clusters with labels, ensuring correspondence between different modality labels. We construct the loss function by incorporating distinctiveness loss and multiplicity loss, both of which are determined by the similarity of neighboring features in the predicted feature space and the difference between distant features. This approach enables efficient feature clustering and cluster class assignment to occur concurrently. Eight UDA cross-modality person ReID experiments are conducted on three real datasets and six synthetic datasets. Experimental results unequivocally demonstrate that the proposed model outperforms the existing state-of-the-art algorithms to a significant degree. Notably, in RegDB → RegDB_light, the Rank-1 accuracy exhibits a remarkable improvement of 8.24%.
- Open Access
Slimmable Compressive Autoencoders for Practical Neural Image Compression
Neural image compression leverages deep neural networks to outperform traditional image codecs in rate-distortion performance. However, the resulting models are also heavy, computationally demanding and generally optimized for a single rate, limiting their practical use. Focusing on practical image compression, we propose slimmable compressive autoencoders (SlimCAEs), where rate (R) and distortion (D) are jointly optimized for different capacities. Once trained, encoders and decoders can be executed at different capacities, leading to different rates and complexities. We show that a successful implementation of SlimCAEs requires suitable capacity-specific RD tradeoffs. Our experiments show that SlimCAEs are highly flexible models that provide excellent rate-distortion performance, variable rate, and dynamic adjustment of memory, computational cost and latency, thus addressing the main requirements of practical image compression.
Open AccessSlimmable Compressive Autoencoders for Practical Neural Image Compression
Neural image compression leverages deep neural networks to outperform traditional image codecs in rate-distortion performance. However, the resulting models are also heavy, computationally demanding and generally optimized for a single rate, limiting their practical use. Focusing on practical image compression, we propose slimmable compressive autoencoders (SlimCAEs), where rate (R) and distortion (D) are jointly optimized for different capacities. Once trained, encoders and decoders can be executed at different capacities, leading to different rates and complexities. We show that a successful implementation of SlimCAEs requires suitable capacity-specific RD tradeoffs. Our experiments show that SlimCAEs are highly flexible models that provide excellent rate-distortion performance, variable rate, and dynamic adjustment of memory, computational cost and latency, thus addressing the main requirements of practical image compression.
- 31 January 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 27 (10709908) , 331-335
Variable Rate Deep Image Compression With Modulated Autoencoder
Variable rate is a requirement for flexible and adaptable image and video compression. However, deep image compression methods (DIC) are optimized for a single fixed rate-distortion (R-D) tradeoff. While this can be addressed by training multiple models for different tradeoffs, the memory requirements increase proportionally to the number of models. Scaling the bottleneck representation of a shared autoencoder can provide variable rate compression with a single shared autoencoder. However, the R-D performance using this simple mechanism degrades in low bitrates, and also shrinks the effective range of bitrates. To address these limitations, we formulate the problem of variable R-D optimization for DIC, and propose modulated autoencoders (MAEs), where the representations of a shared autoencoder are adapted to the specific R-D tradeoff via a modulation network. Jointly training this modulated autoencoder and the modulation network provides an effective way to navigate the R-D operational curve. Our experiments show that the proposed method can achieve almost the same R-D performance of independent models with significantly fewer parameters.
- 31 January 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 27 (10709908) , 331-335
Variable Rate Deep Image Compression With Modulated Autoencoder
Variable rate is a requirement for flexible and adaptable image and video compression. However, deep image compression methods (DIC) are optimized for a single fixed rate-distortion (R-D) tradeoff. While this can be addressed by training multiple models for different tradeoffs, the memory requirements increase proportionally to the number of models. Scaling the bottleneck representation of a shared autoencoder can provide variable rate compression with a single shared autoencoder. However, the R-D performance using this simple mechanism degrades in low bitrates, and also shrinks the effective range of bitrates. To address these limitations, we formulate the problem of variable R-D optimization for DIC, and propose modulated autoencoders (MAEs), where the representations of a shared autoencoder are adapted to the specific R-D tradeoff via a modulation network. Jointly training this modulated autoencoder and the modulation network provides an effective way to navigate the R-D operational curve. Our experiments show that the proposed method can achieve almost the same R-D performance of independent models with significantly fewer parameters.
- 8 May 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 26 (6) , 943-947
Sparse Data Interpolation Using the Geodesic Distance Affinity Space
In this paper, we adapt the geodesic distance-based recursive filter [15] to the sparse data interpolation problem. The proposed technique is general and can be easily applied to any kind of sparse data. We demonstrate the superiority over other interpolation techniques in three experiments for qualitative and quantitative evaluation. In addition, we compare our method with the popular interpolation algorithm presented in the EpicFlow optical flow paper [19] that is intuitively motivated by a similar geodesic distance principle. The comparison shows that our algorithm is more accurate and considerably faster than the EpicFlow interpolation technique.
- 8 May 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 26 (6) , 943-947
Sparse Data Interpolation Using the Geodesic Distance Affinity Space
In this paper, we adapt the geodesic distance-based recursive filter [15] to the sparse data interpolation problem. The proposed technique is general and can be easily applied to any kind of sparse data. We demonstrate the superiority over other interpolation techniques in three experiments for qualitative and quantitative evaluation. In addition, we compare our method with the popular interpolation algorithm presented in the EpicFlow optical flow paper [19] that is intuitively motivated by a similar geodesic distance principle. The comparison shows that our algorithm is more accurate and considerably faster than the EpicFlow interpolation technique.
- Open Access
Sparse data interpolation using the geodesic distance affinity space
In this paper, we adapt the geodesic distance-based recursive filter to the sparse data interpolation problem. The proposed technique is general and can be easily applied to any kind of sparse data. We demonstrate the superiority over other interpolation techniques in three experiments for qualitative and quantitative evaluation. In addition, we compare our method with the popular interpolation algorithm presented in the EpicFlow optical flow paper that is intuitively motivated by a similar geodesic distance principle. The comparison shows that our algorithm is more accurate and considerably faster than the EpicFlow interpolation technique.
Open AccessSparse data interpolation using the geodesic distance affinity space
In this paper, we adapt the geodesic distance-based recursive filter to the sparse data interpolation problem. The proposed technique is general and can be easily applied to any kind of sparse data. We demonstrate the superiority over other interpolation techniques in three experiments for qualitative and quantitative evaluation. In addition, we compare our method with the popular interpolation algorithm presented in the EpicFlow optical flow paper that is intuitively motivated by a similar geodesic distance principle. The comparison shows that our algorithm is more accurate and considerably faster than the EpicFlow interpolation technique.
- 14 January 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 28 (6) , 2936-2947
One-View Occlusion Detection for Stereo Matching With a Fully Connected CRF Model
In this paper, we extend the standard belief propagation (BP) sequential technique proposed in the tree-reweighted sequential method [15] to the fully connected CRF models with the geodesic distance affinity. The proposed method has been applied to the stereo matching problem. Also a new approach to the BP marginal solution is proposed that we call one-view occlusion detection (OVOD). In contrast to the standard winner takes all (WTA) estimation, the proposed OVOD solution allows to find occluded regions in the disparity map and simultaneously improve the matching result. As a result we can perform only one energy minimization process and avoid the cost calculation for the second view and the left-right check procedure. We show that the OVOD approach considerably improves results for cost augmentation and energy minimization techniques in comparison with the standard one-view affinity space implementation. We apply our method to the Middlebury data set and reach state-of-the-art especially for median, average and mean squared error metrics.
- 14 January 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 28 (6) , 2936-2947
One-View Occlusion Detection for Stereo Matching With a Fully Connected CRF Model
In this paper, we extend the standard belief propagation (BP) sequential technique proposed in the tree-reweighted sequential method [15] to the fully connected CRF models with the geodesic distance affinity. The proposed method has been applied to the stereo matching problem. Also a new approach to the BP marginal solution is proposed that we call one-view occlusion detection (OVOD). In contrast to the standard winner takes all (WTA) estimation, the proposed OVOD solution allows to find occluded regions in the disparity map and simultaneously improve the matching result. As a result we can perform only one energy minimization process and avoid the cost calculation for the second view and the left-right check procedure. We show that the OVOD approach considerably improves results for cost augmentation and energy minimization techniques in comparison with the standard one-view affinity space implementation. We apply our method to the Middlebury data set and reach state-of-the-art especially for median, average and mean squared error metrics.
- Open Access
One-view occlusion detection for stereo matching with a fully connected CRF model
In this paper, we extend the standard belief propagation (BP) sequential technique proposed in the tree-reweighted sequential method to the fully connected CRF models with the geodesic distance affinity. The proposed method has been applied to the stereo matching problem. Also a new approach to the BP marginal solution is proposed that we call one-view occlusion detection (OVOD). In contrast to the standard winner takes all (WTA) estimation, the proposed OVOD solution allows to find occluded regions in the disparity map and simultaneously improve the matching result. As a result we can perform only one energy minimization process and avoid the cost calculation for the second view and the left-right check procedure. We show that the OVOD approach considerably improves results for cost augmentation and energy minimization techniques in comparison with the standard one-view affinity space implementation. We apply our method to the Middlebury data set and reach state-of-the-art especially for median, average and mean squared error metrics.
Open AccessOne-view occlusion detection for stereo matching with a fully connected CRF model
In this paper, we extend the standard belief propagation (BP) sequential technique proposed in the tree-reweighted sequential method to the fully connected CRF models with the geodesic distance affinity. The proposed method has been applied to the stereo matching problem. Also a new approach to the BP marginal solution is proposed that we call one-view occlusion detection (OVOD). In contrast to the standard winner takes all (WTA) estimation, the proposed OVOD solution allows to find occluded regions in the disparity map and simultaneously improve the matching result. As a result we can perform only one energy minimization process and avoid the cost calculation for the second view and the left-right check procedure. We show that the OVOD approach considerably improves results for cost augmentation and energy minimization techniques in comparison with the standard one-view affinity space implementation. We apply our method to the Middlebury data set and reach state-of-the-art especially for median, average and mean squared error metrics.