Bruno Zatt

• PhD on Microelectronics
• Professor at Federal University of Pelotas

This chapter presents in detail the four high-throughput architectures developed for the Intra-frame prediction. The first architecture consists of a Depth Intra Skip (DIS) hardware design that adopts a distortion metric replacing, and data reuse in order to obtain energy efficiency.

In March 2011, a call for proposals (ISO/IEC, 2011) was disclosed in parallel to the HEVC standardization aiming to develop a 3D video-coding standard capable of working with an arbitrary number of views, when maintaining high efficiency in compression and image quality. Among several proposals, one proposal based on the HEVC standard and the conce...

This chapter presents in detail the two high-throughput architectures developed for the Inter-frames and Inter-view predictions. The first architecture consists of a Motion/Disparity Estimation system with run-time adaptive memory hierarchy that proposes several complexity-reduction heuristics to reduce energy and memory requirements. The second ha...

Given that the 3D-HEVC extension exploits the MVD concept, and the MVD concept can be considered state of the art in terms of an approach to deal with 3D-video coding, bringing innovations along with its depth-map processing and novel encoding tools, an investigation of the 3D-HEVC encoding tools and the development of algorithms and hardware archi...

Background The HEAD study focuses on collecting an extensive dataset from various tau‐PET tracers, aiming to establish robust anchor values, which are essential for harmonizing tau‐PET measurements. Here, we aim to showcase the capability of converting 3D tau‐PET images into a common scale using the Universal Tau‐PET Scale, Uniτ (tau), and to use t...

Background The HEAD study aims to collect a large dataset of multiple tau‐PET tracers to provide robust anchor values for tau‐PET harmonization. Here, we tested the hypothesis that anchoring two tau tracer uptake values using head‐to‐head measurements has the potential to generate an accurate universal tau‐PET scale, named Uniτ(tau). Methods We as...

Background The HEAD study focuses on collecting an extensive dataset from various tau‐PET tracers, aiming to establish robust anchor values, which are essential for harmonizing tau‐PET measurements. Here, we aim to showcase the capability of converting 3D tau‐PET images into a common scale using the Universal Tau‐PET Scale, Uniτ (tau), and to use t...

Background The HEAD study aims to collect a large dataset of multiple tau‐PET tracers to provide robust anchor values for tau‐PET harmonization. Here, we tested the hypothesis that anchoring two tau tracer uptake values using head‐to‐head measurements has the potential to generate an accurate universal tau‐PET scale, named Uniτ(tau). Methods We as...

Background Positron emission tomography (PET) imaging greatly impacted Alzheimer’s disease (AD) research and diagnosis. which makes predicting PET brain imaging alterations using blood data is of high interest. Additionally, integrating PET and omics data can provide new insights into AD pathophysiology. Here, we implemented a module‐based framewor...

The Versatile Video Coding (VVC) standard achieves high compression rates by introducing new encoding tools, such as the Intra Subpartition Prediction (ISP). However, the ISP increases the computational effort necessary to perform the mode decision of the intra prediction step. This paper proposes a fast intra-mode decision solution for the ISP usi...

Lossy video compression introduces artifacts that can degrade the perceived visual quality of the video. Improving the quality of compressed videos involves mitigating these artifacts through filtering techniques. Deep neural network (DNN) models have emerged as powerful tools for this task, demonstrating effectiveness in artifact reduction. Howeve...

This study proposes a new strategy for intra prediction of dense light fields by reinterpreting the problem as image inpaiting and using convolutional neural networks. Multiple architectures and training techniques were evaluated in order to identify the most efficient configuration for performing intra prediction in a video encoder for this type o...

INTRODUCTION Brain glucose hypometabolism, indexed by the fluorodeoxyglucose positron emission tomography ([¹⁸F]FDG‐PET) imaging, is a metabolic signature of Alzheimer's disease (AD). However, the underlying biological pathways involved in these metabolic changes remain elusive. METHODS Here, we integrated [¹⁸F]FDG‐PET images with blood and hippoc...

The processing of Motion Estimation (ME) in current video encoders demands high memory communication due to the variety of supported block sizes and the high number of candidate blocks that must be evaluated. The intense memory access demanded to process UHD 4K videos in real-time results in undesirable high-power dissipation, which is notably prob...

Nowadays, a large variety of applications and Processing Elements (PE) set the computational environments. We can found multiples PEs embedded in one chip, the called Multiprocessor System-on-Chip (MPSoC), trying to attain the best performance-in parameters like time and energy-according to the application's features. For this, we need to know the...

Audiovisual productions currently dominate internet content, accounting for approximately 82% of online traffic in 2022. With the popularization of adaptive video streaming over the internet, transcoding has become an essential task to allow for bitrate adaptation and compatibility between servers and client devices that support various standards a...

Background Deep learning models, particularly convolutional neural networks (CNNs), have shown promise in Alzheimer’s disease (AD) classification using tau PET data. However, limited sample sizes and unharmonized tau tracers present challenges to developing an agnostic tau tracer tool to predict AD using machine learning. Transfer learning, which l...

This paper presents an error resilience evaluation of the intra prediction step in Versatile Video Coding decoders when approximate storage is employed in the Reference Line Buffer (RLB). We present an error injection framework to simulate the use of approximate storage in the RLB buffer with commonly used bit error rate (BER) values from literatur...

The Symposium on ICs and Systems Design (SBCCI) is an international forum dedicated to ICs and systems design, test, and EDA, held annually in Brazil. Over the past four decades, SBCCI has established itself as an important international forum for the presentation of research results on leading-edge aspects of ICs and systems, such as analog circui...

Vídeos comprimidos geralmente sofrem com efeitos visuais que prejudicam a qualidade percebida pelo usuário. Atualmente, diferentes arquiteturas de aprendizado profundo têm se mostrado eficientes para o problema de melhoria de qualidade em vídeos. No entanto, a maioria delas é treinada e validada usando vídeos gerados por um único padrão de codifica...

This paper proposes a dedicated hardware architecture for the Low-Frequency Non-Separable Transform (LFNST) of the Versatile Video Coding (H.266/VVC) standard. The VVC defines two stages of transformation, where the first stage uses traditional transform types (e.g. DCT-II, DCT-VII and DST-VII), while the secondary transform stage applies the LFNST...

The recent research effort aiming to provide a royalty-free video format resulted in AOMedia Video 1 (AV1), which was launched in 2018. AV1 was developed by the Alliance for Open Media (AOMedia), which groups several major technology companies such as Google, Netflix, Apple, Samsung, Intel, and many others. AV1 is currently one of the most prominen...

The Alliance for Open Media (AOMedia) (AV1) was released in 2018 as a royalty-free and open-source video codec. AV1 was developed by the AOMedia that is composed of many leading tech companies. AV1 has the goal to process ultrahigh definition (UHD) 8K (7680 $\times$ 4320 pixels) and 4K videos (3840 $\times$ 2160 pixels) and to achieve high codi...

This paper presents an error resilience evaluation of the intra prediction step in Versatile Video Coding (VVC) decoders when approximate storage is employed in the Reference Line Buffer (RLB). We present an error injection framework to simulate the use of approximate storage in the RLB buffer with commonly used Bit Error Rate (BER) values from lit...

The diagnosis of Alzheimer’s disease (AD) has been greatly improved due to the fundamental role of positron emission tomography (PET) imaging. Also, predicting PET brain imaging alterations using blood‐based biomarkers is of high interest. This way, integrating PET and omics data can provide new insights into AD pathophysiology. Here, we aimed to d...

Background Positron emission tomography (PET) imaging has greatly improved the diagnosis and monitoring of Alzheimer’s disease (AD). The recently developed neuroinformatic field is expanding analytical and computational strategies to study multimodal neuroscience data. One approach is integrating PET imaging and omics to provide new insights into A...

Modern video encoders like the AOMedia Video 1 (AV1) implement several complex tools to allow the required high level of compression efficiency. The Fractional Motion Estimation (FME) is one of these complex tools, and AV1 FME defines 42 different interpolation filters. To handle such complexity, hardware acceleration using approximate computing ha...

The emerging of 3D video-capable embedded mobile devices is expected due to the popularization of multimedia services and the demand for novel immersive video technologies. Such devices require efficient hardware-friendly heuristics to deal with strict processing requirements and limited energy supply. To contribute to these requirements, this work...

Network congestion and packet loss pose an ever-increasing challenge to video streaming. Despite the research efforts toward making video encoding schemes resilient to lossy network conditions, forwarding devices have not considered monitoring packet content to prioritize packets and minimize the impact of packet loss on video transmission. In this...

This article describes a framework to speed up the HEVC encoding decisions for on-demand transrating of bitstreams. The methods proposed collect information from a high-quality reference bitstream which after processing is used to limit the number of modes evaluated in subsequent re-encodings at different bitrates. In this way, the time required to...

After the AOMedia Video 1 (AV1) bitstream specification was launched by the Alliance for Open Media, the need for converting legacy content encoded with the state-of-the-art High Efficiency Video Coding (HEVC) standard to the new format has arisen in several scenarios. However, transcoding is a complex task composed of a decoding and an encoding pr...

Background Changes in soluble amyloid-beta (Aβ) levels in cerebrospinal fluid (CSF) are detectable at early preclinical stages of Alzheimer’s disease (AD). However, whether Aβ levels can predict downstream AD pathological features in cognitively unimpaired (CU) individuals remains unclear. With this in mind, we aimed at investigating whether a comb...

Background Positron emission tomography (PET) imaging plays a key role in the diagnosis of Alzheimer’s disease (AD). The idea of integrating PET and omics data can provide blood‐new insights about AD pathophysiology. In this context, state‐of‐the‐art techniques such as transcriptomics hold much potential for discovering altered biological processes...

The encoding of 360 videos presents many challenges. These spherical videos require a planar projection stage prior to encoding, and there are no studies on how projection interferes with the encoder decisions. Also, due to the increased resolution, encoding 360 videos is computationally costlier than conventional videos. In this work, we perform a...

Background: Changes in soluble amyloid-beta (Aβ) levels in cerebrospinal fluid (CSF) are detectable at early preclinical stages of Alzheimer's disease (AD). However, whether Aβ levels can predict downstream AD pathological features in cognitively unimpaired (CU) individuals remains unclear. With this in mind, we aimed at investigating whether a com...

The encoding of 360 videos presents many challenges. These videos require a planar projection stage before encoding, and there are no studies on how projection interferes with the encoder decisions. Also, due to the increased resolution, encoding 360 videos is computationally costlier than conventional videos. In this work, we perform an extensive...

This work presents an energy-efficient NoC-based system for real-time multimedia applications employing approximate computing. The proposed video processing system, called SApp-NoC, is efficient in both energy and quality (QoS), employing a scalable NoC architecture composed of processing elements designed to accelerate the HEVC Fractional Motion E...

This paper presents FastInter360, a fast inter mode decision algorithm for accelerating the encoding of ERP 360 videos. The development of FastInter360 involves an in-depth and comprehensive set of evaluations performed to understand the differences in the encoder’s behavior when encoding 360 and conventional videos. These evaluations showed that d...

The digital video coding process imposes severe pressure on memory traffic, leading to considerable power consumption related to frequent DRAM accesses. External off-chip memory demand needs to be minimized by clever architecture/algorithm co-design, thus saving energy and extending battery lifetime during video encoding. To exploit temporal redund...

Approximate computing techniques exploit the characteristics of error-tolerant applications either to provide faster implementations of their computational structures or to achieve substantial improvements in terms of energy efficiency. In video encoding, the motion estimation (ME) stage, including the Integer ME (IME) and the Fractional ME (FME) s...

This article presents an extensive review of the state-of-the-art system-level solutions featuring complexity reduction and/or dedicated hardware designs for the AV1 and VVC video coding formats. These formats introduced several novel coding techniques compared to their predecessors to improve the coding efficiency at the cost of a significant comp...

Video coding applications demand high computational effort to achieve high compression rates at a low perceptual quality expense. In order to reach acceptable encoding time for such applications, modern video coding standards have been em-ploying parallelism approaches to exploit multiprocessing plat-forms, such as the tiling tool from HEVC standar...

Background Positron emission tomography (PET) imaging has been playing a fundamental role in diagnosis of Alzheimer’s disease (AD). Also, in the context of AD, blood‐based biomarkers that are capable of predicting PET brain imaging findings are of high interest. Both PET imaging and transcriptomics are rich sources of different biological informati...

This paper presents an extensive evaluation of the HEVC parallel video coding when using Tiles. The evaluation consists on finding the tiling pattern that yields the maximum possible speedup for a set of video sequences considering several encoding parameters, measuring the coding efficiency variation of using such tiling pattern instead of the uni...

Real-time digital video coding became a mandatory feature in current consumer electronic devices due to the popularization of video applications. However, efficiently encoding videos is an extremely processing/energy-demanding task, especially at high resolutions and frame rates. Thus, the limited energy resources and the dynamically varying system...

This paper presents a complexity and compression efficiency assessment of the 3D-High Efficiency Video Coding (3D-HEVC) encoder. An experimental evaluation was carried out to identify the encoding tools with the highest impact on the computational complexity of 3D-HEVC. The evaluated tools were firstly divided into three categories: HEVC, multiview...

The most common operation to maintain various versions of the same video sequence under different bitrates in live and streaming video applications is video transrating. The typical transrating operation is comprised of a decoding and a complete encoding step in sequence, which results in long processing times and high energy consumption. This work...

This paper presents a memory assessment of the next-generation Versatile Video Coding (VVC). The memory analyses are performed adopting as a baseline the state-of-the-art High-Efficiency Video Coding (HEVC). The goal is to offer insights and observations of how critical the memory requirements of VVC are aggravated, compared to HEVC. The adopted me...

Video encoding draws high research interest, due to the enormous demand for video traffic and real-time encoding for transmission. In video encoding standards such as HEVC (High-Efficiency Video Coding), the final step of the encoding stage is the CABAC (Context-Adaptive Binary Arithmetic Coding). The coding efficiency of the CABAC comes at the cos...

The 3D-HEVC is the state-of-art encoder for 3D videos. This article focuses on the 3D-HEVC depth-map intra prediction. Novel heuristics to reduce the computational effort are explored to design a high-throughput architecture, which supports both the 3D-HEVC tools and the tools inherited from the HEVC. The synthesis results were generated targeting...

With the recent rise in artificial intelligence, several areas of computer science have begun to explore the use of machine learning algorithms. Among them, recent video encoding techniques exploit in large-scale machine learning algorithms, either to aid in compression efficiency or to reduce computational cost. However, this is not a simple task,...