Guilherme Correa
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
Guilherme Correa has a total of 111 co-authors. Here are the top co-authors by number of shared publications:
- Luciano Agostini
UFPEL
50 shared publications - Daniel Palomino
UFPEL
33 shared publications Bruno Zatt
23 shared publications- Luis Alberto da Silva Cruz
Instituto de Telecomunicaes, Coimbra, Portugal
23 shared publications Luciano Agostini
19 shared publications- Pedro Assuncao
Instituto de Telecomunicações
18 shared publications
Latest Publications
- Open Access
- 19 July 2026
- Published by Sociedade Brasileira de Computação
- p. 1046-1051
Two-Step Predictive Models for Low-Complexity Multiple Transform Selection in VVC
The Versatile Video Coding (VVC) standard achieves high compression efficiency but with significantly increased computational complexity. A major contributor is the Multiple Transform Selection (MTS) process, which exhaustively evaluates multiple transform combinations. This work proposes a machine learning approach using lightweight decision trees to replace the exhaustive MTS search. The models are trained with balanced sampling, feature selection, and hyperparameter optimization. Experiments in the All Intra configuration show an average encoding time reduction of 18.76% with only a 0.67% increase in BD-rate, reaching up to 30.90% time savings in the transform stage.
- 19 July 2026
- Published by Sociedade Brasileira de Computação
- p. 1046-1051
Open AccessTwo-Step Predictive Models for Low-Complexity Multiple Transform Selection in VVC
The Versatile Video Coding (VVC) standard achieves high compression efficiency but with significantly increased computational complexity. A major contributor is the Multiple Transform Selection (MTS) process, which exhaustively evaluates multiple transform combinations. This work proposes a machine learning approach using lightweight decision trees to replace the exhaustive MTS search. The models are trained with balanced sampling, feature selection, and hyperparameter optimization. Experiments in the All Intra configuration show an average encoding time reduction of 18.76% with only a 0.67% increase in BD-rate, reaching up to 30.90% time savings in the transform stage.
- Open Access
- 19 July 2026
- Published by Sociedade Brasileira de Computação
- p. 494-505
Perceptual Analysis of Compressed Dynamic Point Clouds in Virtual Reality Environments
This study investigated the perceived quality of dynamic point clouds in virtual reality (VR) under different compression scenarios through an immersive subjective experiment. Six compression scenarios were analyzed, including one employing a machine learning-based solution for codec complexity reduction. A VR application, developed according to ITU-T recommendations, was used to conduct the experiments and collect user scores, response times, and visual attention data. The results indicated agreement between most objective and perceived quality metrics, with moderate variability across participants. Geometric metrics, particularly p2plane and p2point, showed stronger association with MOS than bitrate and luminance PSNR, while heatmaps revealed greater gaze concentration in central and visually informative regions of the point clouds. These findings support the development of perceptually optimized compression and rendering strategies for immersive point cloud applications.
- 19 July 2026
- Published by Sociedade Brasileira de Computação
- p. 494-505
Open AccessPerceptual Analysis of Compressed Dynamic Point Clouds in Virtual Reality Environments
This study investigated the perceived quality of dynamic point clouds in virtual reality (VR) under different compression scenarios through an immersive subjective experiment. Six compression scenarios were analyzed, including one employing a machine learning-based solution for codec complexity reduction. A VR application, developed according to ITU-T recommendations, was used to conduct the experiments and collect user scores, response times, and visual attention data. The results indicated agreement between most objective and perceived quality metrics, with moderate variability across participants. Geometric metrics, particularly p2plane and p2point, showed stronger association with MOS than bitrate and luminance PSNR, while heatmaps revealed greater gaze concentration in central and visually informative regions of the point clouds. These findings support the development of perceptually optimized compression and rendering strategies for immersive point cloud applications.
- Open Access
- 12 April 2026
- Published by Journal of Integrated Circuits and Systems in Journal of Integrated Circuits and Systems
- Vol. 21 (1) , 1-11
Accelerating Versatile Video Coding via Decision-Tree-Based Fast Affine Motion Estimation
As the demand for video transmission surges in remote work, education, and streaming services, the need for continuous advancements in video encoding technologies becomes increasingly evident. Adapting to the evolving demands of efficient video delivery and consumption requires continuous advancement and refinement of video encoding standards. In this scenario, the Versatile Video Coding (VVC) stands out as a leading example, representing the current state-of-the-art in video coding. This work presents a Machine Learning (ML) approach to accelerate one of the novel tools within the VVC: the Affine Motion Estimation (AME). AME is part of the VVC Inter-Frame Prediction, and it is one of the most computationally intensive tools within VVC. The hardware-friendly Decision Tree ML method was explored in this work to accelerate the AME. A total of 12 Decision Trees were trained to determine whether to skip or not to skip the AME for each block size supported by AME in VVC. The proposed approach achieved an average reduction of 5.54% in the total VVC encoding time and an average reduction of 63.20% in the AME execution time, resulting in an average BD-BR efficiency loss of only 0.50%. These results demonstrate that the strategy of using Machine Learning to reduce the computational cost of VVC has the potential to continue yielding significant results in future solutions.
- 12 April 2026
- Published by Journal of Integrated Circuits and Systems in Journal of Integrated Circuits and Systems
- Vol. 21 (1) , 1-11
Open AccessAccelerating Versatile Video Coding via Decision-Tree-Based Fast Affine Motion Estimation
As the demand for video transmission surges in remote work, education, and streaming services, the need for continuous advancements in video encoding technologies becomes increasingly evident. Adapting to the evolving demands of efficient video delivery and consumption requires continuous advancement and refinement of video encoding standards. In this scenario, the Versatile Video Coding (VVC) stands out as a leading example, representing the current state-of-the-art in video coding. This work presents a Machine Learning (ML) approach to accelerate one of the novel tools within the VVC: the Affine Motion Estimation (AME). AME is part of the VVC Inter-Frame Prediction, and it is one of the most computationally intensive tools within VVC. The hardware-friendly Decision Tree ML method was explored in this work to accelerate the AME. A total of 12 Decision Trees were trained to determine whether to skip or not to skip the AME for each block size supported by AME in VVC. The proposed approach achieved an average reduction of 5.54% in the total VVC encoding time and an average reduction of 63.20% in the AME execution time, resulting in an average BD-BR efficiency loss of only 0.50%. These results demonstrate that the strategy of using Machine Learning to reduce the computational cost of VVC has the potential to continue yielding significant results in future solutions.
- Open Access
- 3 February 2026
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 85 (2) , 123
Quantitative assessment of inter-frame prediction in versatile video coding standard
Versatile Video Coding (VVC) is the latest video coding standard established by ISO and ITU-T, boasting a doubled coding efficiency compared to previous standards. However, this substantial enhancement in coding efficiency comes at the cost of an increase in computational demands, posing challenges for practical implementations. This paper conducts a thorough and quantitative assessment of VVC inter-frame prediction, the most computationally intensive tool in the VVC encoder. A comprehensive set of experiments on inter-frame prediction is presented, with a detailed evaluation of the primary bottlenecks of this tool. To the best of the authors’ knowledge, this work is the first in the literature to offer this in-depth assessment of VVC inter-frame prediction.
- 3 February 2026
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 85 (2) , 123
Open AccessQuantitative assessment of inter-frame prediction in versatile video coding standard
Versatile Video Coding (VVC) is the latest video coding standard established by ISO and ITU-T, boasting a doubled coding efficiency compared to previous standards. However, this substantial enhancement in coding efficiency comes at the cost of an increase in computational demands, posing challenges for practical implementations. This paper conducts a thorough and quantitative assessment of VVC inter-frame prediction, the most computationally intensive tool in the VVC encoder. A comprehensive set of experiments on inter-frame prediction is presented, with a detailed evaluation of the primary bottlenecks of this tool. To the best of the authors’ knowledge, this work is the first in the literature to offer this in-depth assessment of VVC inter-frame prediction.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Circuits and Systems Magazine
- Vol. 26 (3) , 81-84
16th IEEE CASS Latin American Symposium on Circuits and Systems (LASCAS 2025) [CASS Conference Highlights]
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Circuits and Systems Magazine
- Vol. 26 (3) , 81-84
16th IEEE CASS Latin American Symposium on Circuits and Systems (LASCAS 2025) [CASS Conference Highlights]
- Open Access
- 10 November 2025
- Published by Sociedade Brasileira de Computação
- p. 560-564
Performance and Energy Consumption Evaluation of the VVC ALF Classification Step Across Different Programming Paradigms and Hardware Platforms
The Adaptive Loop Filter (ALF) in the Versatile Video Coding (VVC) standard enhances visual quality but introduces a significant computational burden, posing a challenge for its implementation on resource-constrained devices. This paper presents a comprehensive evaluation of the ALF classification step, analyzing its execution time and energy consumption across multiple programming paradigms (scalar, SIMD, CUDA) on both a high-performance desktop and an embedded platform. Results show that on the high-performance desktop, SIMD optimization provides a substantial speedup over the scalar baseline with lower energy consumption, while the discrete GPU’s performance is limited by data transfer overhead. Conversely, the embedded system demonstrates a different landscape: its ARM CPU offers superior energy efficiency when compared to the desktop, and its integrated GPU’s runtime is dominated by kernel execution rather than data transfers. These findings underscore that the optimal ALF implementation is highly platform-dependent, hinging on the specific design trade-offs between processing speed and energy efficiency.
- 10 November 2025
- Published by Sociedade Brasileira de Computação
- p. 560-564
Open AccessPerformance and Energy Consumption Evaluation of the VVC ALF Classification Step Across Different Programming Paradigms and Hardware Platforms
The Adaptive Loop Filter (ALF) in the Versatile Video Coding (VVC) standard enhances visual quality but introduces a significant computational burden, posing a challenge for its implementation on resource-constrained devices. This paper presents a comprehensive evaluation of the ALF classification step, analyzing its execution time and energy consumption across multiple programming paradigms (scalar, SIMD, CUDA) on both a high-performance desktop and an embedded platform. Results show that on the high-performance desktop, SIMD optimization provides a substantial speedup over the scalar baseline with lower energy consumption, while the discrete GPU’s performance is limited by data transfer overhead. Conversely, the embedded system demonstrates a different landscape: its ARM CPU offers superior energy efficiency when compared to the desktop, and its integrated GPU’s runtime is dominated by kernel execution rather than data transfers. These findings underscore that the optimal ALF implementation is highly platform-dependent, hinging on the specific design trade-offs between processing speed and energy efficiency.
- Open Access
- 10 November 2025
- Published by Sociedade Brasileira de Computação
- p. 585-597
AI-Based Approaches for Brazilian Sign Language Recognition: A Systematic Literature Review: Insights on Methods, Metrics, and Resources for LIBRAS Recognition
This Systematic Literature Review (SLR) investigates the application of Artificial Intelligence (AI), specifically with Computer Vision (CV) and Natural Language Processing (NLP) techniques for Brazilian Sign Language (LIBRAS) recognition, translation, and interpretation. Following the PRISMA protocol and using the Parsifal tool to support protocol definition and study management, 77 studies were initially identified. After removing duplicates and applying exclusion criteria, 23 studies were selected for in-depth analysis. The review addresses three research questions: (i) AI-based solutions currently employed for LIBRAS recognition, (ii) reported performance metrics and evaluation results, and (iii) datasets used for training and testing, along with their characteristics and limitations. The findings highlight advances in deep learning architectures and computer vision pipelines, while revealing persistent challenges such as limited integration of non-manual features, scarcity of large-scale and multimodal datasets, difficulties in continuous sign recognition, heterogeneous annotation practices, and inadequacy of common bilingual evaluation metrics for capturing syntactic and semantic differences between LIBRAS and Portuguese. Moreover, deployment-related factors, including inference time and computational cost, remain underexplored. These insights provide guidance for future research toward more accurate, efficient, and inclusive LIBRAS recognition systems.
- 10 November 2025
- Published by Sociedade Brasileira de Computação
- p. 585-597
Open AccessAI-Based Approaches for Brazilian Sign Language Recognition: A Systematic Literature Review: Insights on Methods, Metrics, and Resources for LIBRAS Recognition
This Systematic Literature Review (SLR) investigates the application of Artificial Intelligence (AI), specifically with Computer Vision (CV) and Natural Language Processing (NLP) techniques for Brazilian Sign Language (LIBRAS) recognition, translation, and interpretation. Following the PRISMA protocol and using the Parsifal tool to support protocol definition and study management, 77 studies were initially identified. After removing duplicates and applying exclusion criteria, 23 studies were selected for in-depth analysis. The review addresses three research questions: (i) AI-based solutions currently employed for LIBRAS recognition, (ii) reported performance metrics and evaluation results, and (iii) datasets used for training and testing, along with their characteristics and limitations. The findings highlight advances in deep learning architectures and computer vision pipelines, while revealing persistent challenges such as limited integration of non-manual features, scarcity of large-scale and multimodal datasets, difficulties in continuous sign recognition, heterogeneous annotation practices, and inadequacy of common bilingual evaluation metrics for capturing syntactic and semantic differences between LIBRAS and Portuguese. Moreover, deployment-related factors, including inference time and computational cost, remain underexplored. These insights provide guidance for future research toward more accurate, efficient, and inclusive LIBRAS recognition systems.
- Open Access
- 10 November 2025
- Published by Sociedade Brasileira de Computação
- p. 491-495
4D-STDF: Compressed Video Quality Enhancement with 3D Spatio-Temporal Fusion and Deformable Convolution
Compressed videos frequently present artifacts that compromise visual quality. Deep learning models have demonstrated significant effectiveness in mitigating such distortions. In this study, we introduce 3D-STDF, an architecture based on the well-known Spatio-Temporal Deformable Fusion (STDF) and augmented with 3D convolutions to more effectively model temporal dependencies across video frames. Furthermore, we refine the Quality Enhancement (QE) module by integrating residual blocks, thereby enabling the extraction and representation of more intricate spatial features. Experimental results indicate that models based on the 3D-STDF architecture achieved an overall average improvement of up to 0.607 dB in PSNR, clearly outperforming previous STDF-based solutions.
- 10 November 2025
- Published by Sociedade Brasileira de Computação
- p. 491-495
Open Access4D-STDF: Compressed Video Quality Enhancement with 3D Spatio-Temporal Fusion and Deformable Convolution
Compressed videos frequently present artifacts that compromise visual quality. Deep learning models have demonstrated significant effectiveness in mitigating such distortions. In this study, we introduce 3D-STDF, an architecture based on the well-known Spatio-Temporal Deformable Fusion (STDF) and augmented with 3D convolutions to more effectively model temporal dependencies across video frames. Furthermore, we refine the Quality Enhancement (QE) module by integrating residual blocks, thereby enabling the extraction and representation of more intricate spatial features. Experimental results indicate that models based on the 3D-STDF architecture achieved an overall average improvement of up to 0.607 dB in PSNR, clearly outperforming previous STDF-based solutions.
- Open Access
- 10 November 2025
- Published by Sociedade Brasileira de Computação
- p. 446-454
Traditional vs. Neural Video Codecs: Compression Efficiency, Visual Artifacts, and Quality Analysis Beyond PSNR
Video compression technology is rapidly evolving, with end-to-end learned Neural Video Codecs (NVCs) emerging as powerful alternatives to traditional, block-based standards. This paper presents a comparative analysis of two traditional codecs, HEVC and VVC, against two leading NVCs, DCVC-FM and DCVC-RT. We conduct a thorough rate distortion analysis using multiple objective metrics (PSNR, SSIM, VMAF, and LPIPS) and introduce a formal method to identify rate-matched operating points for fair visual comparison. Our results demonstrate that NVCs offer substantial gains not only in signal fidelity but also in perceptual quality, achieving up to 37.82% bitrate savings over HEVC. Furthermore, our analysis reveals fundamental differences in distortion patterns: NVCs excel at preserving structural and color fidelity, producing visually pleasing results free of blocking artifacts, but can sometimes smooth over fine textures. Conversely, traditional codecs are prone to blockiness but can occasionally retain more high-frequency detail. These findings confirm the superior efficiency of NVCs and highlight the need for evaluation methodologies that look beyond PSNR.
- 10 November 2025
- Published by Sociedade Brasileira de Computação
- p. 446-454
Open AccessTraditional vs. Neural Video Codecs: Compression Efficiency, Visual Artifacts, and Quality Analysis Beyond PSNR
Video compression technology is rapidly evolving, with end-to-end learned Neural Video Codecs (NVCs) emerging as powerful alternatives to traditional, block-based standards. This paper presents a comparative analysis of two traditional codecs, HEVC and VVC, against two leading NVCs, DCVC-FM and DCVC-RT. We conduct a thorough rate distortion analysis using multiple objective metrics (PSNR, SSIM, VMAF, and LPIPS) and introduce a formal method to identify rate-matched operating points for fair visual comparison. Our results demonstrate that NVCs offer substantial gains not only in signal fidelity but also in perceptual quality, achieving up to 37.82% bitrate savings over HEVC. Furthermore, our analysis reveals fundamental differences in distortion patterns: NVCs excel at preserving structural and color fidelity, producing visually pleasing results free of blocking artifacts, but can sometimes smooth over fine textures. Conversely, traditional codecs are prone to blockiness but can occasionally retain more high-frequency detail. These findings confirm the superior efficiency of NVCs and highlight the need for evaluation methodologies that look beyond PSNR.