Article
Biochemical Research Methods
Hue Reardon, Anney Che, Brian T. Luke, Sarangan Ravichandran, Jack R. Collins, Uma S. Mudunuri
Summary: AVIA is a web application that allows users to annotate and visualize genomic variant data, investigate functional significance of genetic alterations, and conduct comparative analysis across samples, genes, and pathways. The latest version, AVIA 3.0, offers more filtering options and new services, allowing for greater flexibility in data management.
Article
Multidisciplinary Sciences
Tarek Helmy, Fahim Djatmiko
Summary: This paper presents an automatic framework for semantic annotation of images, utilizing convolutional neural networks to extract image features and recurrent neural networks to process surrounding text and generate annotation sentences. Experimental results show promising performance and comparability to recent image annotation systems.
ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Huanqiang Zeng, Hailiang Huang, Junhui Hou, Jiuwen Cao, Yongtao Wang, Kai-Kuang Ma
Summary: This paper presents a full-reference VQA model designed for assessing the perceptual quality of screen content videos. By extracting spatiotemporal features and employing an adaptive fusion scheme, it achieves more accurate quality assessment results in line with HVS perception.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2022)
Article
Engineering, Electrical & Electronic
Chang Choi, Tian Wang, Christian Esposito, Brij Bhooshan Gupta, Kyungroul Lee
Summary: This paper discusses the definition of object movement and spatio-temporal relations in multimedia data, explores the semantic gap between low-level data and high-level information, and introduces ontology mapping as a method to bridge this gap.
IEEE SENSORS JOURNAL
(2021)
Article
Computer Science, Software Engineering
Alan Lundgard, Arvind Satyanarayan
Summary: This article introduces a model for evaluating the semantic content of natural language descriptions of visualizations, developed through analysis of thousands of sentences and spanning four levels of semantic content. A mixed-methods evaluation with blind and sighted readers reveals significant differences in which semantic content they rank as most useful. The research shows that access to meaningful information is strongly reader-specific.
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
(2022)
Article
Computer Science, Information Systems
Laura Lima Dias, Eduardo Barrere, Jairo Francisco de Souza
Summary: The increase in videos available in educational content repositories has made searching difficult, leading to the use of recommendation systems to help students and teachers find relevant content. This article analyzes the impact of semantic annotation techniques on text extracted from video lecture speech, showing that topic models yield good results in this scenario. Furthermore, a new benchmark for this task has been established for researchers to evaluate new techniques.
JOURNAL OF INFORMATION SCIENCE
(2021)
Article
Chemistry, Multidisciplinary
Eleonora Bernasconi, Davide Di Pierro, Domenico Redavid, Stefano Ferilli
Summary: This paper introduces SKATEBOARD, a tool designed to facilitate knowledge exploration through semantic technologies. It provides users with a clear view of data relationships and dependencies using graph-based exploration and integrates recommendation systems and reasoning capabilities to enhance knowledge discovery. By empowering users to make informed decisions and uncover valuable insights, SKATEBOARD introduces a serendipity effect through its interface exploration.
APPLIED SCIENCES-BASEL
(2023)
Article
Multidisciplinary Sciences
Xiaohan Shu, Ruizhong Yuan, Boying Zheng, Zhizhi Wang, Xiqian Ye, Pu Tang, Xuexin Chen
Summary: In this study, a high-quality genome assembly of M. manilae at the chromosome level was provided using ONT long-read, MGI-SEQ short-read, and Hi-C sequencing methods. The assembled genome size was 282.85 Mb, with 268.17 Mb assigned to 11 pseudochromosomes. The genome contained 152.37 Mb of repetitive elements, representing 53.87% of the total genome size. A total of 15,689 protein-coding genes were predicted, and functional annotations were provided for 13,580 genes. The high-quality genome of M. manilae will serve as a valuable genomic resource for future research on parasitoid wasps.
Article
Computer Science, Information Systems
Houjie Li, Wei Li, Hongda Zhang, Xin He, Mingxiao Zheng, Haiyu Song
Summary: Automatic image annotation, especially based on nearest neighbor models, has become increasingly important in image understanding and pattern recognition. This paper proposes a novel annotation model based on three-pass KNN to address challenges such as semantic gap, label-imbalance, wider range labels, and weak-labeling. Experimental results on benchmark datasets demonstrate the superior performance of the proposed method in comparison to state-of-the-art approaches.
Article
Food Science & Technology
Matej Petkovic, Gorjan Popovski, Barbara Korousic Seljak, Dragi Kocev, Tome Eftimov
Summary: DIETHUB is an AI workflow methodology that annotates online-published recipes or self-reported meals with food concepts, showing high predictive power and correct annotations. The research demonstrates that DIETHUB can successfully analyze corpora of food-related textual documents and provide insight into human dietary behavior.
TRENDS IN FOOD SCIENCE & TECHNOLOGY
(2021)
Article
Biochemical Research Methods
Hong-Dong Li, Changhuo Yang, Zhimin Zhang, Mengyun Yang, Fang-Xiang Wu, Gilbert S. Omenn, Jianxin Wang
Summary: IsoResolve is a novel approach for isoform function prediction that leverages gene function prediction models with domain adaptation to improve performance. It treats gene-level and isoform-level features as source and target domains respectively, and uses domain adaptation to project them into a latent variable space for more accurate predictions.
Article
Engineering, Electrical & Electronic
Xue Song, Baohan Xu, Yu-Gang Jiang
Summary: This study proposes an innovative method for video-in-video advertising using multimodal modeling, which focuses on extracting different representations and learning their complementarity to find content similarity between videos and ads. Experimental results demonstrate the effectiveness and user-friendliness of this method.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2021)
Article
Computer Science, Information Systems
Muhammad Rafiq, Ghazala Rafiq, Gyu Sang Choi
Summary: The rapid expansion and emergence of deep learning in the field of video description have brought various proposals and concerns. Existing deep learning methods show superior computing capabilities and performance but heavily rely on the nature, diversity, and quantity of data.
Article
Computer Science, Information Systems
Hanli Wang, Pengjie Tang, Qinyu Li, Meng Cheng
Summary: Translating a video into natural language is a fundamental yet challenging task, and recent research has focused on this area, achieving state-of-the-art results. However, emotions are often overlooked in video descriptions, resulting in dull and colorless sentences. In this study, a new dataset with emotion expression for video description is constructed, and a fact transfer based framework is proposed to generate sentences with emotion expression. Additionally, a novel approach for sentence evaluation, balancing facts and emotions, is introduced.
IEEE TRANSACTIONS ON MULTIMEDIA
(2022)
Article
Computer Science, Information Systems
Xuguang Zhang, Xin Wei, Liang Zhou, Yi Qian
Summary: This article proposes a self-organized D2D collaborative video content sharing framework for the current challenges in controlling massive devices and transmitting large-volume video data in IoVT. By introducing social attributes and collaboration mechanism, IoVT devices collaborate automatically and share video content through D2D links, offloading the burden. Additionally, a collaborative video streaming strategy is developed by integrating scalable-high-efficiency-video-coding and D2D networking, reducing the impact of network instability on video services.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Telecommunications
Roberto Modonesi, Marco Dalai, Pierangelo Migliorati, Riccardo Leonardi
IEEE COMMUNICATIONS LETTERS
(2020)
Article
Computer Science, Theory & Methods
Alessandro Gnutti, Fabrizio Guerrini, Nicola Adami, Pierangelo Migliorati, Riccardo Leonardi
Summary: This paper explicitly analyzes the performance effects of various wavelet families in compression and denoising tasks, highlighting the significant impact of wavelet parameters. The study serves as a valuable benchmark for future research in this area.
MULTIDIMENSIONAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Multidisciplinary Sciences
Marika Premoli, Daniele Baggi, Marco Bianchetti, Alessandro Gnutti, Marco Bondaschi, Andrea Mastinu, Pierangelo Migliorati, Alberto Signoroni, Riccardo Leonardi, Maurizio Memo, Sara Anna Bonini
Summary: The study proposed a method for automatic classification of ultrasonic vocalizations (USVs) using supervised learning. By utilizing manually built datasets and algorithms like Convolutional Neural Network, it showed that analyzing the full time/frequency information of the USVs spectrograms led to higher performance compared to considering a subset of numerical features.
Article
Computer Science, Artificial Intelligence
Alberto Signoroni, Mattia Savardi, Sergio Benini, Nicola Adami, Riccardo Leonardi, Paolo Gibellini, Filippo Vaccher, Marco Ravanelli, Andrea Borghesi, Roberto Maroldi, Davide Farina
Summary: This study presents an end-to-end deep learning architecture for predicting the degree of lung compromise in COVID-19 patients on Chest X-ray images. The proposed semi-quantitative scoring system shows significant prognostic value and outperforms human annotators. The BS-Net demonstrates high accuracy and self-attentive behavior, showcasing its potential for computer-assisted monitoring in clinical settings.
MEDICAL IMAGE ANALYSIS
(2021)
Article
Engineering, Electrical & Electronic
Khalil Khan, Rehan Ullah Khan, Riccardo Leonardi, Pierangelo Migliorati, Sergio Benini
Summary: This paper surveys the development of head pose estimation methods in both constrained and unconstrained conditions over the past decade. It highlights the ongoing open research topic in this field and summarizes the advantages and disadvantages of existing algorithms, while also suggesting promising directions for future research.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2021)
Article
Computer Science, Information Systems
Marcello Zanardelli, Fabrizio Guerrini, Riccardo Leonardi, Nicola Adami
Summary: In recent years, there has been a proliferation of fake and altered images due to the availability and ease of use of image editing tools. This paper conducts a survey of the latest image forgery detection methods based on Deep Learning (DL) techniques, focusing on copy-move and splicing attacks. The survey discusses the key aspects of these methods, the datasets used for training and validation, as well as their performance. The paper also addresses future research trends and directions in deep learning architectures and evaluation approaches, as well as dataset building for easy methods comparison.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Engineering, Electrical & Electronic
Fabrizio Guerrini, Alessandro Gnutti, Riccardo Leonardi
Summary: This paper explains a unique representation method for describing any finite-energy signal, which consists of an ordered set of positions and a sparse set of signals. By designing and implementing an iterative decomposition approach using mirror operations, the method achieves maximum decoupling and sparsity in the signal representation. Experimental simulations demonstrate the superior approximation capabilities of this approach, making it potentially useful in various domains such as approximation and coding.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Alessandro Gnutti, Fabrizio Guerrini, Riccardo Leonardi, Antonio Ortega
Summary: The Symmetry-Based Graph Fourier Transforms (SBGFTs) have shown promising performance in energy compaction and improvement of HEVC intra coding, making them a potential central role in image compression. Comparisons with other orthogonal transforms such as KLT and SOT further confirm the superior representation ability of SBGFTs in multiple transforms and non-linear approximation perspective.
2021 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Alessandro Gnutti, Fabrizio Guerrini, Riccardo Leonardi
Summary: This paper presents a new 2D transform called Discrete Mirror Transform (DMT), which decomposes a signal into even and odd parts around an optimal location to maximize signal energy split. An optimized version of DMT (ODMT) is also introduced, showing superior performance compared to DCT and DWT in terms of non-linear approximation when applied on images.
2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR)
(2021)
Article
Computer Science, Artificial Intelligence
Alessandro Gnutti, Fabrizio Guerrini, Riccardo Leonardi
Summary: This work addresses the challenging problem of reflection symmetry detection in unconstrained environments by proposing a two-stage solution, which establishes a better correspondence between the outcomes of the algorithm and a human-constructed ground truth, achieves significant performance gains compared to recent symmetry detection competitions, and further validates the approach through perceptual validation experiments with users on a newly built dataset.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Andrea Migliorati, Attilio Fiandrotti, Gianluca Francini, Riccardo Leonardi
Summary: LDVS is a learnable binary local descriptor designed for matching natural images within the MPEG CDVS framework. Through experiments, LDVS descriptors have shown favorable performance in image patch matching and have a moderate parameters count for operations on mobile devices.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Proceedings Paper
Marine & Freshwater Biology
Rossana Sanfilippo, A. Guido, G. Insacco, C. Deias, G. Catania, A. Reitano, R. Leonardi, A. Rosso
13TH INTERNATIONAL POLYCHAETE CONFERENCE (IPC13)
(2020)
Article
Computer Science, Theory & Methods
Fabrizio Guerrini, Marco Dalai, Riccardo Leonardi
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2020)
Proceedings Paper
Acoustics
Fabrizio Guerrini, Alessandro Gnutti, Riccardo Leonardi
2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)
(2019)
Article
Multidisciplinary Sciences
Sergio Benini, Khalil Khan, Riccardo Leonardi, Massimo Mauro, Pierangelo Migliorati