scholarly journals Human Behavior Analysis: A Survey on Action Recognition

2021 ◽  
Vol 11 (18) ◽  
pp. 8324
Author(s):  
Bruno Degardin ◽  
Hugo Proença
Keyword(s):  
Visual Recognition ◽  
Environmental Heterogeneity ◽  
Performance Comparison ◽  
Action Classification ◽  
Human Actions ◽  
Video Images ◽  
Active Research ◽  
Learning Architectures ◽  
Human Behavior Analysis ◽  
Research Domain

The visual recognition and understanding of human actions remain an active research domain of computer vision, being the scope of various research works over the last two decades. The problem is challenging due to its many interpersonal variations in appearance and motion dynamics between humans, without forgetting the environmental heterogeneity between different video images. This complexity splits the problem into two major categories: action classification, recognising the action being performed in the scene, and spatiotemporal action localisation, concerning recognising multiple localised human actions present in the scene. Previous surveys mainly focus on the evolution of this field, from handcrafted features to deep learning architectures. However, this survey presents an overview of both categories and respective evolution within each one, the guidelines that should be followed and the current benchmarks employed for performance comparison between the state-of-the-art methods.

scholarly journals Targeted Contrast Agents for Molecular MRI

Inorganics ◽  
2018 ◽  
Vol 6 (4) ◽  
pp. 129
Author(s):  
Sara Lacerda
Keyword(s):  
Contrast Agents ◽  
Rational Design ◽  
Molecular Mri ◽  
New Class ◽  
Imaging Probes ◽  
Targeted Contrast Agents ◽  
Active Research ◽  
Magnetic Resonance Imaging Mri ◽  
Research Domain

Molecular magnetic resonance imaging (MRI) provides information non-invasively at cellular and molecular levels, for both early diagnosis and monitoring therapeutic follow-up. This imaging technique requires the development of a new class of contrast agents, which signal changes (typically becomes enhanced) when in presence of the cellular or molecular process to be evaluated. Even if molecular MRI has had a prominent role in the advances in medicine over the past two decades, the large majority of the developed probes to date are still in preclinical level, or eventually in phase I or II clinical trials. The development of novel imaging probes is an emergent active research domain. This review focuses on gadolinium-based specific-targeted contrast agents, providing rational design considerations and examples of the strategies recently reported in the literature.

scholarly journals Recent Advances in Mussel-Inspired Synthetic Polymers as Marine Antifouling Coatings

Coatings ◽  
2020 ◽  
Vol 10 (7) ◽  
pp. 653 ◽  
Author(s):  
Ioannis Manolakis ◽  
Usaid Azhar
Keyword(s):  
Amino Acid Derivative ◽  
Self Healing ◽  
Underwater Adhesion ◽  
Marine Biofouling ◽  
Recent Advances ◽  
Antifouling Coatings ◽  
The Common ◽  
Active Research ◽  
Research Domain ◽  
Marine Antifouling

Synthetic oligomers and polymers inspired by the multifunctional tethering system (byssus) of the common mussel (genus Mytilus) have emerged since the 1980s as a very active research domain within the wider bioinspired and biomimetic materials arena. The unique combination of strong underwater adhesion, robust mechanical properties and self-healing capacity has been linked to a large extent to the presence of the unusual α-amino acid derivative l-DOPA (l-3,4-dihydroxyphenylalanine) as a building block of the mussel byssus proteins. This paper provides a short overview of marine biofouling, discussing the different marine biofouling species and natural defenses against these, as well as biomimicry as a concept investigated in the marine antifouling context. A detailed discussion of the literature on the Mytilus mussel family follows, covering elements of their biology, biochemistry and the specific measures adopted by these mussels to utilise their l-DOPA-rich protein sequences (and specifically the ortho-bisphenol (catechol) moiety) in their benefit. A comprehensive account is then given of the key catechol chemistries (covalent and non-covalent/intermolecular) relevant to adhesion, cohesion and self-healing, as well as of some of the most characteristic mussel protein synthetic mimics reported over the past 30 years and the related polymer functionalisation strategies with l-DOPA/catechol. Lastly, we review some of the most recent advances in such mussel-inspired synthetic oligomers and polymers, claimed as specifically aimed or intended for use in marine antifouling coatings and/or tested against marine biofouling species.

scholarly journals On Deep Research Problems in Deep Learning

2021 ◽  
Author(s):  
Anwaar Ulhaq
Keyword(s):  
Deep Learning ◽  
Visual Recognition ◽  
Exploratory Research ◽  
Research Issues ◽  
Learning Success ◽  
Research Problems ◽  
Technical Issues ◽  
Active Research ◽  
The One ◽  
Research Frontiers

The subject of deep learning has emerged in the last decade as one of the most promising approaches to machine learning. Today, certainly, much of the recent progress in artificial intelligence is due to it, but research challenges are still unresolved and remain open to the research community. This paper attempts to offer a comprehensive review of deep learning progress in active research frontiers. On the one side, by presenting a brief overview of deep learning success, we inspire researchers to work in deep learning. On the other hand, we examine a range of technical issues, and open research issues that we believe are relevant topics for exploratory research. As deep learning applies to various fields, we restrict this paper’s scope to visual recognition tasks to analyze these problems with a specific lens. However, these problems will be broadly applicable to other fields. It will make it easier for new researchers to recognize outstanding research problems in the deep learning domain.

scholarly journals ConvNet and LSH-Based Visual Localization Using Localized Sequence Matching

Sensors ◽  
2019 ◽  
Vol 19 (11) ◽  
pp. 2439 ◽  
Author(s):  
Yongliang Qiao ◽  
Cindy Cappelle ◽  
Yassine Ruichek ◽  
Tao Yang
Keyword(s):  
Visual Recognition ◽  
Distance Matrix ◽  
Performance Comparison ◽  
Locality Sensitive Hashing ◽  
Visual Localization ◽  
Data Sets ◽  
Sequence Matching ◽  
Real World Data ◽  
Convolutional Network ◽  
Illumination Changes

Convolutional Network (ConvNet), with its strong image representation ability, has achieved significant progress in the computer vision and robotic fields. In this paper, we propose a visual localization approach based on place recognition that combines the powerful ConvNet features and localized image sequence matching. The image distance matrix is constructed based on the cosine distance of extracted ConvNet features, and then a sequence search technique is applied on this distance matrix for the final visual recognition. To speed up the computational efficiency, the locality sensitive hashing (LSH) method is applied to achieve real-time performances with minimal accuracy degradation. We present extensive experiments on four real world data sets to evaluate each of the specific challenges in visual recognition. A comprehensive performance comparison of different ConvNet layers (each defining a level of features) considering both appearance and illumination changes is conducted. Compared with the traditional approaches based on hand-crafted features and single image matching, the proposed method shows good performances even in the presence of appearance and illumination changes.

QSAR/QSPR Model Research of Complicated Samples

2013 ◽  
Vol 740 ◽  
pp. 306-309
Author(s):  
Yan Kun Li ◽  
Xiao Ying Ma
Keyword(s):  
Research Methods ◽  
Quantitative Structure ◽  
Qspr Model ◽  
Structure Activity ◽  
Model Research ◽  
Qspr Study ◽  
Chemical Informatics ◽  
Active Research ◽  
Research Domain

QSAR/QSPR study is a hot issue in present chemical informatics research, and is the very active research domain. In present, a large number of QSAR/QSPR (quantitative structure-activity/property relationships) models have been widely studied and applied in a lot of different areas. This paper overviews the developments, research methods and applications of QSAR/QSPR model.

scholarly journals Implementation of facial emotion recognition using cnn on jetson TX2

2021 ◽  
pp. 11-18
Author(s):  
Pham Minh Quyen ◽  
Phung Thanh Huy ◽  
Do Duy Tan ◽  
Huynh Hoang Ha ◽  
Truong Quang Phuc
Keyword(s):  
Neural Network ◽  
Complex Structure ◽  
Computation Time ◽  
Performance Comparison ◽  
Facial Emotion ◽  
Laptop Computer ◽  
Basic Emotions ◽  
Testing Dataset ◽  
Learning Architectures ◽  
Extensive Performance

In this paper, a convolutional neural network (CNN), one of the most popular deep learning architectures used for facial extraction research, has been implemented on NVIDIA Jetson TX2 hardware. Different from many existing approaches investigating CNN with complex structure and large parameters, we have focused on building a robust neural network through extensive performance comparison and evaluation. In addition, we have collected a dataset using a built-in camera on a laptop computer. Specifically, we have applied our model on Jetson TX2 hardware to take advantage of the computational power of the embedded GPU to optimize computation time and data training. In particular, both FER2013 and RAF datasets with seven basic emotions have been used for training and testing purposes. Finally, the evaluation results show that the proposed method achieves an accuracy of up to 72% on the testing dataset.

Conspecific Recognition in Pigeons (Columba Livia) Using Dynamic Video Images

Behaviour ◽  
1998 ◽  
Vol 135 (1) ◽  
pp. 43-53 ◽  
Author(s):  
Keyword(s):  
Visual Recognition ◽  
Columba Livia ◽  
The Body ◽  
Courtship Display ◽  
Display Duration ◽  
Video Images ◽  
Conspecific Recognition ◽  
Courtship Behaviors ◽  
Courtship Displays

AbstractVideo images of pigeons were used to examine the degree to which these images are equivalent to real live conspecifics by analyzing the natural behaviors of pigeons in the presence of each stimulus. Three aspects of courtship display (i.e. bowing, tail-dragging, and vocalizations) were selected and the display duration for each was measured. When videotaped images of female pigeons were presented as stimuli, the display duration by male pigeons was not significantly different from that for the live birds. In contrast, the subjects showed much shorter, or no, displays to the video images of a non-pigeon bird (cockatoo) and an empty chamber. The results suggest that the video images of pigeons contained necessary information to trigger the courtship behaviors. Furthermore, the present study examined which features of the video images were critical for triggering the displays by manipulating the images. Thus, the subjects' behaviors were more vigorous (1) when video images were in motion rather than still, and (2) when the head-only region was visible rather than the body-only region. These results suggest that motion and facial/head characteristics are important features. Collectively, the results indicate that the use of video images as stimuli and courtship displays as measures provide a useful method to study the visual recognition of conspecifics in birds.

scholarly journals Recognition of COVID-19 from CT Scans Using Two-Stage Deep-Learning-Based Approach: CNR-IEMN

Sensors ◽  
2021 ◽  
Vol 21 (17) ◽  
pp. 5878 ◽  
Author(s):  
Fares Bougourzi ◽  
Riccardo Contino ◽  
Cosimo Distante ◽  
Abdelmalik Taleb-Ahmed
Keyword(s):  
Deep Learning ◽  
Community Acquired Pneumonia ◽  
Research Topic ◽  
The Other ◽  
Ct Scans ◽  
Second Stage ◽  
Volumetric Ct ◽  
Active Research ◽  
Validation Set ◽  
Learning Architectures

Since the appearance of the COVID-19 pandemic (at the end of 2019, Wuhan, China), the recognition of COVID-19 with medical imaging has become an active research topic for the machine learning and computer vision community. This paper is based on the results obtained from the 2021 COVID-19 SPGC challenge, which aims to classify volumetric CT scans into normal, COVID-19, or community-acquired pneumonia (Cap) classes. To this end, we proposed a deep-learning-based approach (CNR-IEMN) that consists of two main stages. In the first stage, we trained four deep learning architectures with a multi-tasks strategy for slice-level classification. In the second stage, we used the previously trained models with an XG-boost classifier to classify the whole CT scan into normal, COVID-19, or Cap classes. Our approach achieved a good result on the validation set, with an overall accuracy of 87.75% and 96.36%, 52.63%, and 95.83% sensitivities for COVID-19, Cap, and normal, respectively. On the other hand, our approach achieved fifth place on the three test datasets of SPGC in the COVID-19 challenge, where our approach achieved the best result for COVID-19 sensitivity. In addition, our approach achieved second place on two of the three testing sets.

Sign in / Sign up

Export Citation Format

Share Document