scholarly journals VideoRecognition - Uma proposta de serviço para reconhecimento de elementos de vídeo em larga escala

2018 ◽  
Author(s):  
Antonio J.G. Busson ◽  
Álan L.V. Guedes ◽  
Gabriel N.P. Dos Santos ◽  
Carlos de Salles Soares Neto ◽  
Ruy Luiz Milidiú ◽  
...  
Keyword(s):  
Deep Learning ◽  
Speech Processing ◽  
The State ◽  
Video Classification ◽  
Multimedia Services ◽  
Traditional Use ◽  
Ongoing Research ◽  
It Usage ◽  
Learning Research ◽  
State Of Art

Deep Learning research has allowed significant advancement of various segments of multimedia, especially in tasks related to speech processing, hearing and computational vision. However, some video services are still focused only on the traditional use of media (capture, storage, transmission and presentation). In this paper, we discuss our ongoing research towards a DLaS, i.e. Deep Learning as a Service. This way, we present the state of art in video classification and recognition. Then we propose the VideoRecognition as DLaS to support the tasks such as: image classification and video scenes, object detection and facial recognition. We discuss the usage of the proposed service in the context of the video@RNP repository. Our main contributions consist on dicussussions over the state of art and it usage in nowdays multimedia services.

scholarly journals Deep Learning for Video Classification: A Review

2021 ◽  
Author(s):  
Atiq Rehman ◽  
Samir Brahim Belhaouari
Keyword(s):  
Deep Learning ◽  
Network Architecture ◽  
Classification Task ◽  
Future Research ◽  
Video Classification ◽  
Learning Models ◽  
Learning Methods ◽  
Research Directions ◽  
Concise Review ◽  
State Of Art

<div><div><div><p>Video classification task has gained a significant success in the recent years. Specifically, the topic has gained more attention after the emergence of deep learning models as a successful tool for automatically classifying videos. In recognition to the importance of video classification task and to summarize the success of deep learning models for this task, this paper presents a very comprehensive and concise review on the topic. There are a number of existing reviews and survey papers related to video classification in the scientific literature. However, the existing review papers are either outdated, and therefore, do not include the recent state-of-art works or they have some limitations. In order to provide an updated and concise review, this paper highlights the key findings based on the existing deep learning models. The key findings are also discussed in a way to provide future research directions. This review mainly focuses on the type of network architecture used, the evaluation criteria to measure the success, and the data sets used. To make the review self- contained, the emergence of deep learning methods towards automatic video classification and the state-of-art deep learning methods are well explained and summarized. Moreover, a clear insight of the newly developed deep learning architectures and the traditional approaches is provided, and the critical challenges based on the benchmarks are highlighted for evaluating the technical progress of these methods. The paper also summarizes the benchmark datasets and the performance evaluation matrices for video classification. Based on the compact, complete, and concise review, the paper proposes new research directions to solve the challenging video classification problem.</p></div></div></div>

scholarly journals Deep Learning for Video Classification: A Review

2021 ◽  
Author(s):  
Atiq Rehman ◽  
Samir Brahim Belhaouari
Keyword(s):  
Deep Learning ◽  
Network Architecture ◽  
Classification Task ◽  
Future Research ◽  
Video Classification ◽  
Learning Models ◽  
Learning Methods ◽  
Research Directions ◽  
Concise Review ◽  
State Of Art

<div><div><div><p>Video classification task has gained a significant success in the recent years. Specifically, the topic has gained more attention after the emergence of deep learning models as a successful tool for automatically classifying videos. In recognition to the importance of video classification task and to summarize the success of deep learning models for this task, this paper presents a very comprehensive and concise review on the topic. There are a number of existing reviews and survey papers related to video classification in the scientific literature. However, the existing review papers are either outdated, and therefore, do not include the recent state-of-art works or they have some limitations. In order to provide an updated and concise review, this paper highlights the key findings based on the existing deep learning models. The key findings are also discussed in a way to provide future research directions. This review mainly focuses on the type of network architecture used, the evaluation criteria to measure the success, and the data sets used. To make the review self- contained, the emergence of deep learning methods towards automatic video classification and the state-of-art deep learning methods are well explained and summarized. Moreover, a clear insight of the newly developed deep learning architectures and the traditional approaches is provided, and the critical challenges based on the benchmarks are highlighted for evaluating the technical progress of these methods. The paper also summarizes the benchmark datasets and the performance evaluation matrices for video classification. Based on the compact, complete, and concise review, the paper proposes new research directions to solve the challenging video classification problem.</p></div></div></div>

Opponent Center-Surround Contrast: Colour to Grey Conversion

2020 ◽  
Vol 2020 (1) ◽  
pp. 105-108
Author(s):  
Ali Alsam
Keyword(s):  
Evolutionary Algorithms ◽  
Three Dimensional ◽  
The State ◽  
Colour Space ◽  
Vision Science ◽  
State Of Art ◽  
Over Time

Vision is the science that informs us about the biological and evolutionary algorithms that our eyes, opticnerves and brains have chosen over time to see. This article is an attempt to solve the problem of colour to grey conversion, by borrowing ideas from vision science. We introduce an algorithm that measures contrast along the opponent colour directions and use the results to combine a three dimensional colour space into a grey. The results indicate that the proposed algorithm competes with the state of art algorithms.

Kidney transplantation in identical twins: the state of art and a case report

2019 ◽  
Vol 21 (4) ◽  
pp. 458-465
Author(s):  
A.I. Sushkov ◽  
◽  
T.A. Astrelina ◽  
E.V. Shestero ◽  
V.A. Nikitina ◽  
...  
Keyword(s):  
Kidney Transplantation ◽  
Case Report ◽  
The State ◽  
Identical Twins ◽  
State Of Art

scholarly journals A Review of Size Engineering-Enabled Electrocatalysts for Li–S Chemistry

2021 ◽  
Author(s):  
Xi Zhang ◽  
Yaping Zhang ◽  
xijun wei ◽  
Chaohui Wei ◽  
Yingze Song
Keyword(s):  
Energy Density ◽  
High Energy ◽  
The State ◽  
High Energy Density ◽  
State Of Art

Li–S batteries (LBSs) have received extensive attention owing to their remarkable theoretical capacity (1672 mA h g–1) and high energy density (2600 Wh kg–1), far beyond the state of art...

scholarly journals Representation Learning for Fine-Grained Change Detection

Sensors ◽  
2021 ◽  
Vol 21 (13) ◽  
pp. 4486
Author(s):  
Niall O’Mahony ◽  
Sean Campbell ◽  
Lenka Krpalkova ◽  
Anderson Carvalho ◽  
Joseph Walsh ◽  
...  
Keyword(s):  
Deep Learning ◽  
Change Detection ◽  
Model Calibration ◽  
State Of The Art ◽  
Representation Learning ◽  
Machine Intelligence ◽  
The State ◽  
Sensor Data ◽  
Fine Grained ◽  
Learning Techniques

Fine-grained change detection in sensor data is very challenging for artificial intelligence though it is critically important in practice. It is the process of identifying differences in the state of an object or phenomenon where the differences are class-specific and are difficult to generalise. As a result, many recent technologies that leverage big data and deep learning struggle with this task. This review focuses on the state-of-the-art methods, applications, and challenges of representation learning for fine-grained change detection. Our research focuses on methods of harnessing the latent metric space of representation learning techniques as an interim output for hybrid human-machine intelligence. We review methods for transforming and projecting embedding space such that significant changes can be communicated more effectively and a more comprehensive interpretation of underlying relationships in sensor data is facilitated. We conduct this research in our work towards developing a method for aligning the axes of latent embedding space with meaningful real-world metrics so that the reasoning behind the detection of change in relation to past observations may be revealed and adjusted. This is an important topic in many fields concerned with producing more meaningful and explainable outputs from deep learning and also for providing means for knowledge injection and model calibration in order to maintain user confidence.

scholarly journals Assessing the Potential of Catch-Only Models to Inform on the State of Global Fisheries and the UN’s SDGs

2021 ◽  
Vol 13 (11) ◽  
pp. 6101
Author(s):  
Rishi Sharma ◽  
Henning Winker ◽  
Polina Levontin ◽  
Laurence Kell ◽  
Dan Ovando ◽  
...  
Keyword(s):  
Stock Assessment ◽  
Fishing Effort ◽  
The State ◽  
Sources Of Information ◽  
Ongoing Research ◽  
Traditional Assessment ◽  
Stock Status ◽  
Stock Assessments ◽  
Recent Developments ◽  
Measure Of Performance

Catch-only models (COMs) have been the focus of ongoing research into data-poor stock assessment methods. Two of the most recent models that are especially promising are (i) CMSY+, the latest refined version of CMSY that has progressed from Catch-MSY, and (ii) SRA+ (Stock Reduction Analysis Plus) a recent developments in field. Comparing COMs and evaluating their relative performance is essential for determining the state of regional and global fisheries that may be lacking necessary data that would be required to run traditional assessment models. In this paper we interrogate how performance of COMs can be improved by incorporating additional sources of information. We evaluate the performance of COMs on a dataset of 48 data-rich ICES (International Council for the Exploration of Seas) stock assessments. As one measure of performance, we consider the ability of the model to correctly classify stock status using FAO’s 3-tier classification that is also used for reporting on sustainable development goals to the UN. Both COMs showed notable bias when run with their inbuilt default heuristics, but as the quality of prior information increased, classification rates for the terminal year improved substantially. We conclude that although further COM refinements show some potential, most promising is the ongoing research into developing biomass or fishing effort priors for COMs in order to be able to reliably track stock status for the majority of the world’s fisheries currently lacking stock assessments.

scholarly journals Synthetic speech detection through short-term and long-term prediction traces

2021 ◽  
Vol 2021 (1) ◽  
Author(s):  
Clara Borrelli ◽  
Paolo Bestagini ◽  
Fabio Antonacci ◽  
Augusto Sarti ◽  
Stefano Tubaro
Keyword(s):  
Deep Learning ◽  
Speech Processing ◽  
Synthetic Speech ◽  
Opinion Formation ◽  
Closed Set ◽  
Speech Detection ◽  
Open Set ◽  
Technological Advances ◽  
Speech Generation ◽  
Long Term Prediction

AbstractSeveral methods for synthetic audio speech generation have been developed in the literature through the years. With the great technological advances brought by deep learning, many novel synthetic speech techniques achieving incredible realistic results have been recently proposed. As these methods generate convincing fake human voices, they can be used in a malicious way to negatively impact on today’s society (e.g., people impersonation, fake news spreading, opinion formation). For this reason, the ability of detecting whether a speech recording is synthetic or pristine is becoming an urgent necessity. In this work, we develop a synthetic speech detector. This takes as input an audio recording, extracts a series of hand-crafted features motivated by the speech-processing literature, and classify them in either closed-set or open-set. The proposed detector is validated on a publicly available dataset consisting of 17 synthetic speech generation algorithms ranging from old fashioned vocoders to modern deep learning solutions. Results show that the proposed method outperforms recently proposed detectors in the forensics literature.

Sign in / Sign up

Export Citation Format

Share Document