MUSED: A multimedia multi-document dataset for topic segmentation

2018 ◽  
Vol 24 (6) ◽  
pp. 921-946
Author(s):  
PEDRO MOTA ◽  
MAXINE ESKENAZI ◽  
LUÍSA COHEUR
Keyword(s):  
State Of The Art ◽  
The State ◽  
Topic Segmentation ◽  
Document Segmentation ◽  
Multiple Documents ◽  
The Impact ◽  
Segmentation Models

AbstractResearch on topic segmentation has recently focused on segmenting documents by taking advantage of documents covering the same topics. In order to properly evaluate such approaches, a dataset of related documents is needed. However, existing datasets are limited in the number of related documents per domain. In addition, most of the available datasets do not consider documents from different media sources (PowerPoints, videos, etc.), which pose specific challenges to segmentation. We fill this gap with the MUltimedia SEgmentation Dataset (MUSED), a collection of documents manually segmented, from different media sources, in seven different domains, with an average of twenty related documents per domain. In this paper, we describe the process of building MUSED. A multi-annotator study is carried out to determine if it is possible to observe agreement among human judges and characterize their disagreement patterns. In addition, we use MUSED to compare the state-of-the-art topic segmentation techniques, including the ones that take advantage of related documents. Moreover, we study the impact of having documents from different media sources in the dataset. To the best of our knowledge, MUSED is the first dataset that allows a straightforward evaluation of both single- and multiple-documents topic segmentation techniques, as well as to study how these behave in the presence of documents from different media sources. Results show that some techniques are, indeed, sensitive to different media sources, and also that current multi-document segmentation models do not outperform previous models, pointing to a research line that needs to be boosted.

scholarly journals Assessing the Impact of the Loss Function, Architecture and Image Type for Deep Learning-Based Wildfire Segmentation

2021 ◽  
Vol 11 (15) ◽  
pp. 7046
Author(s):  
Jorge Francisco Ciprián-Sánchez ◽  
Gilberto Ochoa-Ruiz ◽  
Lucile Rossi ◽  
Frédéric Morandini
Keyword(s):  
Deep Learning ◽  
Loss Function ◽  
State Of The Art ◽  
Fire Detection ◽  
Loss Functions ◽  
Wildfire Spread ◽  
Combine Information ◽  
The Impact ◽  
Image Type ◽  
Segmentation Models

Wildfires stand as one of the most relevant natural disasters worldwide, particularly more so due to the effect of climate change and its impact on various societal and environmental levels. In this regard, a significant amount of research has been done in order to address this issue, deploying a wide variety of technologies and following a multi-disciplinary approach. Notably, computer vision has played a fundamental role in this regard. It can be used to extract and combine information from several imaging modalities in regard to fire detection, characterization and wildfire spread forecasting. In recent years, there has been work pertaining to Deep Learning (DL)-based fire segmentation, showing very promising results. However, it is currently unclear whether the architecture of a model, its loss function, or the image type employed (visible, infrared, or fused) has the most impact on the fire segmentation results. In the present work, we evaluate different combinations of state-of-the-art (SOTA) DL architectures, loss functions, and types of images to identify the parameters most relevant to improve the segmentation results. We benchmark them to identify the top-performing ones and compare them to traditional fire segmentation techniques. Finally, we evaluate if the addition of attention modules on the best performing architecture can further improve the segmentation results. To the best of our knowledge, this is the first work that evaluates the impact of the architecture, loss function, and image type in the performance of DL-based wildfire segmentation models.

Researching Telework

2000 ◽  
pp. 255-276 ◽  
Author(s):  
Nicole B. Ellison
Keyword(s):  
Organizational Culture ◽  
State Of The Art ◽  
The State ◽  
Future Research ◽  
Scholarly Literature ◽  
Home Based ◽  
The Family ◽  
The Individual ◽  
The Impact

This chapter examines the state of the art in telework research. The author reviews the most central scholarly literature examining the phenomenon of telework (also called home-based work or telecommuting) and develops a framework for organizing this body of work. She organizes previous research on telework into six major thematic concerns relating to the definition, measurement, and scope of telework; management of teleworkers; travel-related impacts of telework; organizational culture and employee isolation; boundaries between “home” and “work” and the impact of telework on the individual and the family. Areas for future research are suggested.

scholarly journals (Sub)titles in cinematic virtual reality: a descriptive study

2021 ◽  
Author(s):  
Belén Agulló ◽  
◽  
Anna Matamala ◽  
Keyword(s):  
New York ◽  
Virtual Reality ◽  
State Of The Art ◽  
New York Times ◽  
Descriptive Study ◽  
The State ◽  
Corpus Analysis ◽  
Content Creation ◽  
Standard Practices ◽  
The Impact

Virtual reality has attracted the attention of industry and researchers. Its applications for entertainment and audiovisual content creation are endless. Filmmakers are experimenting with different techniques to create immersive stories. Also, subtitle creators and researchers are finding new ways to implement (sub)titles in this new medium. In this article, the state-of-the-art of cinematic virtual reality content is presented and the current challenges faced by filmmakers when dealing with this medium and the impact of immersive content on subtitling practices are discussed. Moreover, the different studies on subtitles in 360º videos carried out so far and the obtained results are reviewed. Finally, the results of a corpus analysis are presented in order to illustrate the current subtitle practices by The New York Times and the BBC. The results have shed some light on issues such as position, innovative graphic strategies or the different functions, challenging current subtitling standard practices in 2D content.

scholarly journals Impact of Cryogenics on Cavitation through an Orifice: A Review

Energies ◽  
2021 ◽  
Vol 14 (24) ◽  
pp. 8319
Author(s):  
Claudia Esposito ◽  
Johan Steelant ◽  
Maria Rosaria Vetrano
Keyword(s):  
State Of The Art ◽  
The State ◽  
Hydrodynamic Cavitation ◽  
Knowledge Gaps ◽  
Propulsion Systems ◽  
Clear Vision ◽  
Basic Concepts ◽  
The Impact

Cryogenic cavitation affects the operation of liquid propulsion systems during the first phase of a launch. Its effects within orifices or turbopumps can range from mild instabilities to catastrophic damages to the structures, jeopardizing the launch itself. Therefore, to ensure the proper designing of propulsion systems, cavitation phenomena cannot be neglected. Although hydrodynamic cavitation has been studied for decades, the impact of the nature of the fluid has been sparsely investigated. Therefore, this review, beginning from the basic concepts of cavitation, analyzes the literature dedicated to hydrodynamic cryogenic cavitation through an orifice. Our review provides a clear vision of the state-of-the-art from experimental and modeling viewpoints, identifies the knowledge gaps in the literature, and proposes a way to further investigate cryogenic cavitation in aerospace science.

scholarly journals A Comparative Study of Semantic Segmentation Models for Identification of Grape with Different Varieties

Agriculture ◽  
2021 ◽  
Vol 11 (10) ◽  
pp. 997
Author(s):  
Yun Peng ◽  
Aichen Wang ◽  
Jizhan Liu ◽  
Muhammad Faheem
Keyword(s):  
Image Enhancement ◽  
Precision Agriculture ◽  
State Of The Art ◽  
Semantic Segmentation ◽  
Histogram Equalization ◽  
Performance Difference ◽  
Convolutional Network ◽  
Enhancement Method ◽  
The Impact ◽  
Segmentation Models

Accurate fruit segmentation in images is the prerequisite and key step for precision agriculture. In this article, aiming at the segmentation of grape cluster with different varieties, 3 state-of-the-art semantic segmentation networks, i.e., Fully Convolutional Network (FCN), U-Net, and DeepLabv3+ applied on six different datasets were studied. We investigated: (1) the segmentation performance difference of the 3 studied networks; (2) The impact of different input representations on segmentation performance; (3) The effect of image enhancement method to improve the poor illumination of images and further improve the segmentation performance; (4) The impact of the distance between grape clusters and camera on segmentation performance. The experiment results show that compared with FCN and U-Net the DeepLabv3+ combined with transfer learning is more suitable for the task with an intersection over union (IoU) of 84.26%. Five different input representations, namely RGB, HSV, L*a*b, HHH, and YCrCb obtained different IoU, ranging from 81.5% to 88.44%. Among them, the L*a*b got the highest IoU. Besides, the adopted Histogram Equalization (HE) image enhancement method could improve the model’s robustness against poor illumination conditions. Through the HE preprocessing, the IoU of the enhanced dataset increased by 3.88%, from 84.26% to 88.14%. The distance between the target and camera also affects the segmentation performance, no matter in which dataset, the closer the distance, the better the segmentation performance was. In a word, the conclusion of this research provides some meaningful suggestions for the study of grape or other fruit segmentation.

scholarly journals Digital Archiving for Archaeology: the state of the art in Argentina

2021 ◽  
Author(s):  
Andrés D. Izeta ◽  
Roxana Cattáneo
Keyword(s):  
State Of The Art ◽  
National Level ◽  
Legal Framework ◽  
The State ◽  
Digital Data ◽  
Archaeological Research ◽  
General Evaluation ◽  
Research Infrastructures ◽  
The Status ◽  
The Impact

This article discusses the state-of-the art of digital archives for archaeological research in Argentina. It also presents and characterises the national and international legal framework and the role played by funding agencies and professional bodies in archaeological practice. In addition, it reports how legal corpora regulate the impact on the management of archaeological digital data. Research infrastructures available at the national level are described, such as the Suquía, an institutional digital archive devoted to archaeology since 2016. Finally, we make a general evaluation of the status quo of research infrastructures mostly concerned with preserving and disseminating data from archaeological research at the national level.

Analyzing the Digital License Reselling Problem and Its Impact on E-Commerce

2014 ◽  
pp. 577-597
Author(s):  
Tarek Gaber ◽  
Ning Zhang
Keyword(s):  
State Of The Art ◽  
Digital Rights Management ◽  
The State ◽  
Digital Rights ◽  
Rights Management ◽  
The Impact

Existing Digital Rights Management (DRM) systems allow consumers to buy digital licenses to access the corresponding contents on their devices. However, with these DRM systems, the consumers are unable to resell their licenses. Supporting digital license reselling adds additional challenges to DRM technologies and could find a new E-market. The aims of this chapter are as follows. The problem of reselling digital licenses is formally formulated. Then the state-of-the-art of the existing license reselling solutions proposed in the literature is discussed. Their strengths and limitations are analyzed. Then a framework allowing a consumer to resell his/her license to another consumer without compromising the underlying security of the DRM system is proposed. Finally, the impact of allowing license reselling on E-commerce is discussed.

Sarcopenia: Beyond Muscle Atrophy and into the New Frontiers of Opportunistic Imaging, Precision Medicine, and Machine Learning

2018 ◽  
Vol 22 (03) ◽  
pp. 307-322 ◽  
Author(s):  
Leon Lenchik ◽  
Robert Boutin
Keyword(s):  
Public Health ◽  
Machine Learning ◽  
Precision Medicine ◽  
Muscle Atrophy ◽  
State Of The Art ◽  
The State ◽  
Future Directions ◽  
Common Diagnosis ◽  
The Impact

AbstractAs populations continue to age worldwide, the impact of sarcopenia on public health will continue to grow. The clinically relevant and increasingly common diagnosis of sarcopenia is at the confluence of three tectonic shifts in medicine: opportunistic imaging, precision medicine, and machine learning. This review focuses on the state-of-the-art imaging of sarcopenia and provides context for such imaging by discussing the epidemiology, pathophysiology, consequences, and future directions in the field of sarcopenia.

Sign in / Sign up

Export Citation Format

Share Document