“计算机视觉”版本间的差异
来自iCenter Wiki
(→SSD) |
(→对象检测) |
||
第2行: | 第2行: | ||
计算机视觉(Computer Vision),简称CV,包含对象检测、人脸识别、文字识别等。 | 计算机视觉(Computer Vision),简称CV,包含对象检测、人脸识别、文字识别等。 | ||
+ | |||
+ | 计算机视觉任务(Visual Task)包括:分类(Classification)、定位(localization)、 检测(detection)和 分割(segmentation),即识别对象的类别,位置,以及所在场景解析与标记。 | ||
= 对象检测 = | = 对象检测 = | ||
第26行: | 第28行: | ||
# Faster R-CNN Towards real-time object detection with region proposal networks, NIPS, 2015. | # Faster R-CNN Towards real-time object detection with region proposal networks, NIPS, 2015. | ||
+ | |||
==YOLO== | ==YOLO== | ||
第39行: | 第42行: | ||
[https://github.com/balancap/SSD-Tensorflow SSD-Tensorflow] | [https://github.com/balancap/SSD-Tensorflow SSD-Tensorflow] | ||
+ | |||
+ | ==Segmentation== | ||
+ | |||
+ | mask r-CNN | ||
+ | |||
+ | Learning to Segment Every Thing, | ||
= 人脸识别 = | = 人脸识别 = |
2018年2月22日 (四) 10:06的版本
计算机视觉
计算机视觉(Computer Vision),简称CV,包含对象检测、人脸识别、文字识别等。
计算机视觉任务(Visual Task)包括:分类(Classification)、定位(localization)、 检测(detection)和 分割(segmentation),即识别对象的类别,位置,以及所在场景解析与标记。
对象检测
对象检测,Object Detection,是计算机视觉一项基本功能。
R-CNN
Ross Girshick,FAIR研究员,R-CNN 和YOLO 算法的开创者。
R-CNN ((Region-based Convolutional Network)) (Code: Matlab)
- Region based convolutional networks for accurate object detection and segmentation, TPAMI, 2015.
- Rich feature hierarchies for accurate object detection and semantic segmentation, CVPR 2014.
Fast R-CNN (Code: Python)
- Fast R-CNN, ICCV 2015.
Faster R-CNN (Code: Matlab, Python)
- Faster R-CNN Towards real-time object detection with region proposal networks, NIPS, 2015.
YOLO
(Code Yolo)
- You Only Look Once: Unified, Real-Time Object Detection, CVPR 2016
SSD
- Wei Liu, Dragomir Anguelov, Dumitru Erhan, Christian Szegedy, Scott Reed, Cheng-Yang Fu, Alexander C. Berg, SSD: Single Shot MultiBox Detector, ECCV 2016.
Segmentation
mask r-CNN
Learning to Segment Every Thing,
人脸识别
人脸识别,Face Recognition,分为传统机器学习和深度神经网络两大类方法。
传统机器学习
特征脸算法(Eigenface)
局部二值模式(Local Binary Patterns,简称LBP)
Fisherface算法
深度神经网络
CMU - OpenFace