Filters








11 Hits in 9.5 sec

Stingray Detection of Aerial Images Using Augmented Training Images Generated by A Conditional Generative Model [article]

Yi-Min Chou, Chien-Hung Chen, Keng-Hao Liu, Chu-Song Chen
2018 arXiv   pre-print
Unlike traditional data augmentation methods that generate new data only for image classification, our proposed method that mixes foreground and background together can generate new data for an object  ...  In this paper, we present an object detection method that tackles the stingray detection problem based on aerial images.  ...  We introduce a data augmentation method called mixed bg-fg synthesis to fuse background patches and foreground objects without apparent artifacts, which is achieved by a new generative network C-GLO.  ... 
arXiv:1805.04262v3 fatcat:r4depknrhfaazemfyqlim4b3aq

Plankton classification with high-throughput submersible holographic microscopy and transfer learning

Liam MacNeil, Sergey Missan, Junliang Luo, Thomas Trappenberg, Julie LaRoche
2021 BMC Ecology and Evolution  
Background Plankton are foundational to marine food webs and an important feature for characterizing ocean health.  ...  Results Here we use a light and portable digital in-line holographic microscope (The HoloSea) with maximum optical resolution of 1.5 μm, intensity-based object detection through a volume, and four different  ...  JLR is supported by a grant to Dr.  ... 
doi:10.1186/s12862-021-01839-0 pmid:34134620 fatcat:y4h6cwahgbdy7bcqnxxmfplwum

Detecting Biological Locomotion in Video: A Computational Approach [article]

Soo Min Kang, Richard P. Wildes
2021 arXiv   pre-print
An alternative algorithm, based on generic features combined with learning is assembled out of components from allied areas of investigation, also is presented as a basis of comparison.  ...  Quantitative results indicate that the proposed algorithm considerably outperforms the alternative approach, supporting the hypothesis that biolocomotion can be detected reliably based on its distinct  ...  Acknowledgements The authors thank Robert Allison, Michael Brown, Marcus Brubaker, Dorita Chang and Nikolaus Troje for valuable comments on the work presented in this report.  ... 
arXiv:2105.12661v1 fatcat:ag3p6gbn3fhpddgashubjtkase

Sustainable Drainage Systems [chapter]

Neil Berwick
2016 Sustainable Surface Water Management  
The authors also thank Bill Lucas, who is chairing an American Society of Civil Engineer's Task Committee focusing on computational models for use in LID, for his insight.  ...  The authors also want to extend thanks to Østfold University College and The Norwegian University of Life Sciences for financing this study.  ...  , and then changes in flow rates and volume from natural to developed conditions; Oxley and Mays [27] proposed an optimization model, based upon the simulated annealing method, to optimize the size and  ... 
doi:10.1002/9781118897690.ch4 fatcat:3gizoj2cffdnjjekbemzydv7um

ConVRgence (VRIC) Virtual Reality International Conference Proceedings

Simon Richir
2019 International Journal of Virtual Reality  
We also wish to thank Mette Duus Hansen for collaboration on this project.  ...  ACKNOWLEDGEMENTS The authors wish to thank Bo Geng from the Aalborg University Health Department for her help and supervision throughout the project.  ...  Depending on the lens used the effect achieved changes. II. BACKGROUND A. ANATOMY OF A LENTICULAR LENS B.  ... 
doi:10.20870/ijvr.2019..0.2920 fatcat:rzbvdg34q5aj7mpgnedsofd4ze

Intermezzo I [chapter]

Mihnea Tănăsescu
2022 Ecocene Politics  
I therefore acknowledge my unpayable debt to ancestors living and spectral, human and non-human. They are the soil that nurture present and future life.  ...  All have provided indispensable funding for seeing this project through. Mariana și Doru au lăsat mai multe urme decât se pot număra. Darurile lor nu se răsplătesc; se dau mai departe.  ...  Acknowledgments Acknowledgments Giulia made the space necessary for this work to be written, and the time to hear it out all too often.  ... 
doi:10.11647/obp.0274.03 fatcat:phkpaxq2ojditbdslopp2ra7ny

Green speculations: science fiction and transformative environmentalism

2013 ChoiceReviews  
Along with Peter Blaze Corcoran, Maria Roca, and Joe Wisdom, Rebecca and Jim have mentored me for many years and put me on the path that led to this book.  ...  I am especially grateful for the attention that Rebecca Totaro and Jim Wohlpart of Florida Gulf Coast University gave to this project.  ...  references either to recreate the social conditions of a predetermined, univocal social system or to create freeing conditions based on an ecology of private experiences.  ... 
doi:10.5860/choice.50-3693 fatcat:vdk7szzmdfba7ay5tz35nusxla

Deep one-class learning: a deep learning approach to anomaly detection [article]

Lukas Ruff, Technische Universität Berlin, Klaus-Robert Müller, Marius Kloft
2021
This thesis proposes Deep One-Class Learning, a deep learning approach to anomaly detection that is based on the one-class classification paradigm.  ...  The lack of effective solutions in such settings has sparked an interest in developing anomaly detection methods based on deep learning, which has enabled breakthroughs in other machine learning problems  ...  Methods based on one-class classification try to avoid a full estimation of the probability density as an intermediate step to anomaly detection.  ... 
doi:10.14279/depositonce-12250 fatcat:7pvu5nxc3rbw3exz72mquyxz6a

Commercial electronic publishing over open networks: a global approach based on mobile objects (agents)

Jean-Henry Morin, Dionysios Tsichritzis
1999
This represents a major change for network applications thus leading to a new paradigm.  ...  Upon arrival of an agent on a location an init() and a start() methods are called. Before leaving a location, a stop() method is called.  ...  should rely on a standard widely accepted formatting language for electronic content composition and rendering IP-8 information providers should be able to reflect the editorial process within the electronic  ... 
doi:10.13097/archive-ouverte/unige:146922 fatcat:gv2xctfg4jhmjbxi6trwbriqou

Multimodal sensor-based control for human-robot interaction

Andrea Cherubini, Andrea
unpublished
Traditionally, heterogeneous sensor data was fed to fusion algorithms (e.g., Kalman or Bayesian-based), so as to provide state estimation for modeling the environment.  ...  Each of these axes will be presented here, before concluding with a general view of the issues at stake, and on the research projects that I plan to carry out in the upcoming years.  ...  Acknowledgements It would have been impossible for me to accomplish the research work presented in this manuscript without the fundamental help of friends and colleagues, first in Rennes and then in Montpellier  ... 
fatcat:2jjc237fnjdjjonhq672as2zw4

Landmarks: reading the Gove Peninsula [article]

Kitty Eggerking, University, The Australian National, University, The Australian National
2014
To chart a course through the two disparate traditions, much of the focus is on the ways that land is conceptualised.  ...  For instance, it takes into account the instability of the Menzies government in 1963; and, further, it examines the relevant parliamentary standing orders to show that the bark petitions were in order  ...  7 One construction camp was at Wallaby Beach, the other was located close to where the Gove Peninsula Motel now stands.  ... 
doi:10.25911/5d7395882d0e4 fatcat:kwwa27c6cbbgpng7pux6d3w5sq