【CV】图像分割二十年,盘点影响力最大的10篇论文
机器学习初学者
共 5052字,需浏览 11分钟
·
2020-10-01 11:53
极市导读
图像分割(image segmentation)技术是计算机视觉领域的重要的研究方向,近些年,图像分割技术迅猛发展,在多个视觉研究领域都有着广泛的应用。本文盘点了近20年来影响力最大的 10 篇论文。
发布信息: 2017,16th IEEE International Conference on Computer Vision (ICCV)
论文:https://arxiv.org/abs/1703.06870
代码:https://github.com/facebookresearch/Detectron
发布信息:2015,IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
论文:https://arxiv.org/pdf/1511.00561.pdf
代码:https://github.com/aizawan/segnet
发布信息:2018,IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
DeepLabv2:https://arxiv.org/pdf/1606.00915.pdf
DeepLabv3:https://arxiv.org/pdf/1706.05587.pdf
DeepLabv3+:https://arxiv.org/pdf/1802.02611.pdf
代码:https://github.com/tensorflow/models/tree/master/research/deeplab
发布信息:2011,IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
论文和代码:https://www2.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/resources.html
发布信息:2004,INTERNATIONAL JOURNAL OF COMPUTER VISION
论文和代码:http://cs.brown.edu/people/pfelzens/segment/
发布信息:2012,IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
论文和代码:https://ivrlwww.epfl.ch/supplementary_material/RK_SLICSuperpixels/index.html
发布信息:2015,18th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI)
代码:https://lmb.informatik.uni-freiburg.de/people/ronneber/u-net/
发布信息:2002,IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
发布信息:2000,IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 论文:https://ieeexplore.ieee.org/abstract/document/1000236
发布信息:2015,IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
代码:https://github.com/shelhamer/fcn.berkeleyvision.org
(2)增大数据尺寸的反卷积(deconv)层。能够输出精细的结果。
(3)结合不同深度层结果的跳级(skip)结构。同时确保鲁棒性和精确性。
[1]FCN的学习及理解(Fully Convolutional Networks for Semantic Segmentation),CSDN
[2]mean shift 图像分割 (一),CSDN
[3]https://zhuanlan.zhihu.com/p/49512872
[4]图像分割—基于图的图像分割(Graph-Based Image Segmentation),CSDN
[5]https://www.cnblogs.com/fourmi/p/9785377.html
往期精彩回顾
获取一折本站知识星球优惠券,复制链接直接打开:
https://t.zsxq.com/662nyZF
本站qq群704220115。
加入微信群请扫码进群(如果是博士或者准备读博士请说明):
评论