6 Hits in 4.3 sec

MSFD:Multi-Scale Receptive Field Face Detector [article]

Qiushan Guo, Yuan Dong, Yu Guo, Hongliang Bai
2019 arXiv   pre-print
This paper presents our Multi-Scale Receptive Field Face Detector (MSFD), which has superior performance on detecting faces at different scales and enjoys real-time inference speed.  ...  We aim to study the multi-scale receptive fields of a single convolutional neural network to detect faces of varied scales.  ...  Fig. 1 : 1 The network architecture of Multi-Scale Receptive Field Face Detector. Fig. 2 : 2 The Context-Texture module.  ... 
arXiv:1903.04147v1 fatcat:nae3krrsgzgkhpeq5a4ztm7oqi

SE-IYOLOV3: An Accurate Small Scale Face Detector for Outdoor Security

Zhenrong Deng, Rui Yang, Rushi Lan, Zhenbing Liu, Xiaonan Luo
2020 Mathematics  
To further improve the detection performance, we adopt the SENet structure to enhance the global receptive field of the network.  ...  Small scale face detection is a very difficult problem. In order to achieve a higher detection accuracy, we propose a novel method, termed SE-IYOLOV3, for small scale face in this work.  ...  [30] proposed MSFD, which is a multi-scale face detector in the reception domain and can detect faces of different scales.  ... 
doi:10.3390/math8010093 fatcat:krm7m7mynbgrlb3d3n2cqhrzba

Image Matching from Handcrafted to Deep Features: A Survey

Jiayi Ma, Xingyu Jiang, Aoxiang Fan, Junjun Jiang, Junchi Yan
2020 International Journal of Computer Vision  
This survey can serve as a reference for (but not limited to) researchers and engineers in image matching and related fields.  ...  Multiscale sampling or changed receptive field would make these deep learning-based detectors invariant to scale, where the scale or rotation information is directly estimated in networks.  ...  Fixed-Scale Detectors Adaptive-Scale Detectors It is desirable to adaptively fit with the scale in detection.  ... 
doi:10.1007/s11263-020-01359-2 fatcat:a2epfaolwjfm5mcrsmn7g6sd7m

Toward the widespread application of low-cost technologies in coastal ocean observing (Internet of Things for the Ocean)

2021 Mediterranean Marine Science  
They both use unlicensed frequencies, and thus do not incur spectrum usage fees, but are not protected from in-terference from other users and face limitations in transmission times.  ...  Therefore, an observing platform that can send data (hourly, daily, weekly or monthly) and that can be battery powered should consume as little as possible both in transmission and in reception.  ... 
doi:10.12681/mms.25060 fatcat:dmyp3sgoprba5abtn3hdwyz3ra

Environmental impact assessment for the representative site report - Marseille

Brizzi Giulio, Sabbagh Maroua
2022 Zenodo  
Environmental impact assessment of a Multi-purpose platform MPP designed for integrating offshore aquaculture with renewable energy extraction.  ...  Electro-sensitive organisms are able to detect two types of E field: localized polar and larger scale uniform E fields.  ...  The first mode refers to electro-receptive species as Elamobranchs (Sharks, Rays, Skates, and Ratfish). It is generally assumed that they use the iE field detection for navigation.  ... 
doi:10.5281/zenodo.5896547 fatcat:gkm3aml2j5fprf37isg3nembgi

D5.1 – Environmental factors and mapping to pilots

Olivier Le Brun, Deborah Mille, Igor Kegalj, Teodora Milosevic, Stjepan Pilicic, Luka Traven
2021 Zenodo  
Ship waste is a significant item in the recommendations of the Port Reception Facilities Directive and the appropriateness of port reception facilities for the reception of new types of shipwrecks and  ...  Participants are required to provide adequate reception facilities for the produced waste.  ...  facilities Regarding the assessment of the aspect of the bilge water has been assassed in the section of waste in shipping field (ANNEX I_MARPOL 73/78 Convention _Port Reception facilities) A.1.  ... 
doi:10.5281/zenodo.5552724 fatcat:gecx6r46hbgw5adhtlty2zky4i