论文笔记
文章平均质量分 91
clover_my
这个作者很懒,什么都没留下…
展开
-
论文笔记(YOLOv4)-2020-YOLOv4: Optimal Speed and Accuracy of Object Detection
YOLOv4: Optimal Speed and Accuracy of Object DetectionAlexey Bochkovskiy,Chien-Yao Wang (Institute of Information Science, Academia Sinica, Taiwan),Hong-Yuan Mark Liao (Institute of Information Science, Academia Sinica, Taiwan)论文地址:https://arxiv.org/pd翻译 2021-04-19 20:22:06 · 789 阅读 · 0 评论 -
论文笔记(SSD)-2016-SSD: Single Shot MultiBox Detector
论文笔记(SSD)-2016-SSD: Single Shot MultiBox Detector论文:https://arxiv.org/abs/1512.02325源码:https://github.com/weiliu89/caffe/tree/ssd摘要:我们提出了一种利用单一深度神经网络检测图像中目标的方法,称为 SSD。它将 bounding boxes 的输出空间离散...翻译 2020-03-19 14:37:19 · 789 阅读 · 0 评论 -
论文笔记(RF-pose)-2018-Through-Wall Human Pose Estimation Using Radio Signals
论文笔记(RF-pose)论文:Through-Wall Human Pose Estimation Using Radio Signals1、摘要:介绍了一种深度神经网络方法,利用WiFi频率中穿过墙壁并反射到人体的无线电信号(射频,RF),来估计人体的二维姿态。由于无法对无线电信号进行标注,因此使用最先进的视觉模型(OpenPose )来提供跨模式的监督。具体来说,在训练过...原创 2019-04-16 17:25:58 · 3926 阅读 · 0 评论 -
论文笔记-2019-Object Detection in 20 Years: A Survey
Object Detection in 20 Years: A SurveyZhengxia Zou, Zhenwei Shi, Member, IEEE, Yuhong Guo, and Jieping Ye, Senior Member, IEEE论文获取:https://arxiv.org/abs/1905.05055v2目录Object Detection in 20 Year...翻译 2019-06-18 21:08:17 · 32015 阅读 · 12 评论 -
论文笔记(SR)-2017-Super-Resolution via Deep Learning(详解SRCNN)
论文笔记(SR)-Super-Resolution via Deep Learning(详解SRCNN)论文下载:https://arxiv.org/pdf/1706.09077.pdf文章以深度学习为背景,对SR文献进行了考察。关注多媒体的三个重要方面,即图像、视频和多维度,尤其是深度映射。在每种情况下,首先以数据集和最先进的SR方法(不包括深度学习)的形式引入相关基准。接下来是对各...原创 2019-05-23 15:52:21 · 8532 阅读 · 2 评论 -
论文笔记(FCOS)-2019-FCOS:Fully Convolutional One-Stage Object Detection
FCOS:Fully Convolutional One-Stage Object DetectionZhi Tian Chunhua Shen∗ Hao Chen Tong HeThe University of Adelaide, Australia目录FCOS:Fully Convolutional One-Stage Object Detection1. Introduc...翻译 2019-06-21 11:52:41 · 3307 阅读 · 0 评论 -
论文笔记(FoveaBox)-2019-FoveaBox: Beyond Anchor-based Object Detector
FoveaBox: Beyond Anchor-based Object DetectorTao Kong1 Fuchun Sun1 Huaping Liu1 Yuning Jiang2 Jianbo Shi31Department of Computer Science and Technology, Tsinghua University, 2Beijing Natio...翻译 2019-06-24 17:53:41 · 2445 阅读 · 0 评论 -
论文笔记-2019-ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy
ImageNet训练的神经网络偏向于纹理,增加形状偏置可以提高精度ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy论文:https://arxiv.org/abs/1811.12231Github:https://github.com/rgeirhos/te...翻译 2019-08-22 20:22:01 · 3918 阅读 · 0 评论