没有合适的资源?快使用搜索试试~ 我知道了~
首页The Future of Real-Time SLAM and Deep Learning vs SLAM.
资源详情
资源评论
资源推荐
The Future of Real-Time SLAM and Deep Learning vs SLAM
Wednesday, January 13, 2016 (2016-01-13T04:20:00-05:00) andrew davison (http://www.computervisionblog.com/search/label/andrew%20davison) ,
bundle adjustment (http://www.computervisionblog.com/search/label/bundle%20adjustment) , DTAM (http://www.computervisionblog.com/search/label/DTAM) ,
DynamicFusion (http://www.computervisionblog.com/search/label/DynamicFusion) , iccv 2015 (http://www.computervisionblog.com/search/label/iccv%202015) ,
jakob engel (http://www.computervisionblog.com/search/label/jakob%20engel) , KinectFusion (http://www.computervisionblog.com/search/label/KinectFusion) ,
LSD-SLAM (http://www.computervisionblog.com/search/label/LSD-SLAM) , marc pollefeys (http://www.computervisionblog.com/search/label/marc%20pollefeys) ,
pose (http://www.computervisionblog.com/search/label/pose) , PTAM (http://www.computervisionblog.com/search/label/PTAM) , real-time
(http://www.computervisionblog.com/search/label/real-time) , richard newcombe (http://www.computervisionblog.com/search/label/richard%20newcombe) ,
robotics (http://www.computervisionblog.com/search/label/robotics) , segmentation (http://www.computervisionblog.com/search/label/segmentation) , sfm
(http://www.computervisionblog.com/search/label/sfm) , SLAM (http://www.computervisionblog.com/search/label/SLAM) , workshop
(http://www.computervisionblog.com/search/label/workshop) , zisserman (http://www.computervisionblog.com/search/label/zisserman) 27 Comments
Last month's International Conference of Computer Vision (ICCV) was full of Deep Learning (http://www.computervisionblog.com/2015/12/iccv-2015-
twenty-one-hottest-research.html)techniques, but before we declare an all-out ConvNet victory, let's see how the other "non-learning" geometric side of
computer vision is doing. Simultaneous Localization and Mapping, or SLAM, is arguably one of the most important algorithms in Robotics, with pioneering
work done by both computer vision and robotics research communities. Today I'll be summarizing my key points from ICCV'sFuture of Real-Time SLAM
(http://wp.doc.ic.ac.uk/thefutureofslam/programme/)Workshop, which was held on the last day of the conference (December 18th, 2015).
Today's post contains a brief introduction to SLAM,a detailed description of what happened at the workshop (with summaries of all 7 talks),and some take-
home messages from the Deep Learning-focused panel discussion at the end of the session.
(http://1.bp.blogspot.com/-3WNdePDKHQw/VpOAwv91xWI/AAAAAAAAOcY/Q6oXFwf14Jw/s1600/slammies2.png)
SLAM visualizations.Can you identify any of these SLAM algorithms?
Part I: Why SLAM Matters
Visual SLAM algorithms are able to simultaneously build 3D maps of the world while tracking the location and orientation of the camera (hand-held or
head-mounted for AR or mounted on a robot).SLAM algorithms are complementary to ConvNets and Deep Learning: SLAM focuses on geometric problems
and Deep Learning is the master of perception (recognition) problems. If you want a robot to go towards your refrigerator without hitting a wall, use SLAM.
If you want the robot to identify the items inside your fridge, use ConvNets.
(http://openmvg.readthedocs.org/en/latest/_images/structureFromMotion.png)
Basics of SfM/SLAM: From point observation and intrinsic camera parameters, the 3D structure of a scene is computed from the estimated motion of the
camera. For details, seeopenMVG website (http://openmvg.readthedocs.org/en/latest/).
SLAMis a real-time version ofStructurefromMotion (SfM). Visual SLAM or vision-based SLAM is a camera-only variant of SLAM which forgoes expensive
laser sensors andinertial measurement units (IMUs). Monocular SLAM uses a single camera while non-monocular SLAM typically uses a pre-calibrated
fixed-baseline stereo camera rig. SLAM is prime example of a what is called a "Geometric Method" in Computer Vision. In fact, CMU's Robotics Institute
splits the graduate level computer vision curriculum into a Learning-based Methods in Vision (http://graphics.cs.cmu.edu/courses/16-824-
S15/index.html) course and a separate Geometry-Based Methods in Vision (http://www.cs.cmu.edu/~hebert/geom.html) course.
Structure from Motion vs Visual SLAM
Structure from Motion (SfM) and SLAM are solving a very similar problem, but while SfM is traditionally performed in an offline fashion, SLAM has been
slowly moving towards the low-power / real-time / single RGB camera mode of operation. Many of the today’s top experts in Structure from Motion work
for some of the world’s biggest tech companies, helping make maps better. Successful mapping products like Google Maps could not have been built
without intimate knowledge of multiple-view geometry, SfM, and SLAM. A typical SfM problem is the following: given a large collection of photos of a
single outdoor structure (like the Colliseum), construct a 3D model of the structure and determine the camera's poses. The image collection is processed in
an offline setting, and large reconstructions can take anywhere between hours and days.
(http://www.cs.cornell.edu/~snavely/bundler/images/Colosseum.jpg)
SfM Software:Bundler (http://www.cs.cornell.edu/~snavely/bundler/)isone of the most successful SfM open source libraries
Here are some popular SfM-related software libraries:
Bundler (http://www.cs.cornell.edu/~snavely/bundler/),an open-source Structure from Motion toolkit
Libceres (http://ceres-solver.org/), a non-linear least squares minimizer (useful for bundle adjustment problems)
Andrew Zisserman's Multiple-View Geometry MATLAB Functions (http://www.robots.ox.ac.uk/~vgg/hzbook/code/)
Visual SLAM vs Autonomous Driving
While self-driving cars are one of the most important applications of SLAM, according to Andrew Davison, one of the workshop organizers, SLAM for
Autonomous Vehicles deserves its own research track. (And as we'll see, none of the workshop presenters talked about self-driving cars). For many years to
come it will make sense to continue studying SLAM from a research perspective, independent of any single Holy-Grail application. While there are just too
many system-level details and tricks involved with autonomous vehicles, research-grade SLAM systems require very little more than a webcam, knowledge
of algorithms, and elbow grease. As a research topic, Visual SLAM is much friendlier to thousands of early-stage PhD students who’ll first need years of in-
lab experience with SLAM before even starting to think about expensive robotic platforms such as self-driving cars.
(http://spectrum.ieee.org/image/1948541)
Google's Self-Driving Car's perception system. From IEEE Spectrum's "How Google's Self-Driving Car Works
(http://spectrum.ieee.org/automaton/robotics/artificial-intelligence/how-google-self-driving-car-works)"
Related: March 2015 blog post, Mobileye's quest to put Deep Learning inside every new car (http://www.computervisionblog.com/2015/03/mobileyes-
quest-to-put-deep-learning.html).
Related: One way Google's Cars Localize Themselves (http://mappingignorance.org/2014/04/07/one-way-googles-cars-localize/)
Part II: The Future of Real-time SLAM
Now it's time toofficiallysummarize and comment on the presentations from The Future of Real-time SLAM workshop.Andrew Davison
(http://www.doc.ic.ac.uk/~ajd/index.html) started the day with an excellent historical overview of SLAM called 15 years of vision-based SLAM
(http://wp.doc.ic.ac.uk/thefutureofslam/wp-content/uploads/sites/93/2015/12/slides_ajd.pdf), and his slides have good content for an introductory
robotics course.
For those of you who don’t know Andy, he is the one and only Professor Andrew Davison of Imperial College London. Most known for his 2003 MonoSLAM
system, he was one of the first to show how to build SLAM systems from a single “monocular”camera at a time when just everybody thought you needed a
stereo “binocular” camera rig. More recently, his work has influenced the trajectory of companies such as Dyson and the capabilities of their robotic
systems (e.g., the brand new Dyson360 (http://www.computervisionblog.com/2015/05/dyson-360-eye-and-baidu-deep-learning.html)).
I remember Professor Davidson from the Visual SLAM tutorial he gave at the BMVC Conference back in 2007
(http://www.cs.bris.ac.uk/Research/Vision/Realtime/bmvctutorial/). Surprisingly very little has changed in SLAM compared to the rest of the machine-
learning heavy work being done at the main vision conferences. In the past 8 years, object recognition has undergone 2-3 mini revolutions, while today's
SLAM systems don't look much different than they did 8 years ago. The best way to see the progress of SLAM is to take a look at the most successful and
memorable systems.In Davison’s workshop introduction talk, he discussed some of these exemplary systems which were produced by the research
community over the last 10-15 years:
MonoSLAM
PTAM
FAB-MAP
DTAM
KinectFusion
Davison vs Horn: The next chapter in Robot Vision
Davison also mentioned that he is working on a new Robot Vision book, which should be an exciting treat for researchers in computer vision, robotics, and
artificial intelligence. The last Robot Vision book (https://mitpress.mit.edu/books/robot-vision) was written by B.K. Horn (1986), and it’s about time for an
updated take on Robot Vision.
(http://3.bp.blogspot.com/-
Oh94IZlctLA/VpWlSwN_WEI/AAAAAAAAOdw/fKDBj8KQoGM/s1600/robotvision-01.png)
A new robot vision book?
While I’ll gladly read a tome that focuses on the philosophy of robot vision, personally I would like the book to focus on practical algorithms for robot
剩余11页未读,继续阅读
bitiworm
- 粉丝: 1
- 资源: 6
上传资源 快速赚钱
- 我的内容管理 收起
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
会员权益专享
最新资源
- ExcelVBA中的Range和Cells用法说明.pdf
- 基于单片机的电梯控制模型设计.doc
- 主成分分析和因子分析.pptx
- 共享笔记服务系统论文.doc
- 基于数据治理体系的数据中台实践分享.pptx
- 变压器的铭牌和额定值.pptx
- 计算机网络课程设计报告--用winsock设计Ping应用程序.doc
- 高电压技术课件:第03章 液体和固体介质的电气特性.pdf
- Oracle商务智能精华介绍.pptx
- 基于单片机的输液滴速控制系统设计文档.doc
- dw考试题 5套.pdf
- 学生档案管理系统详细设计说明书.doc
- 操作系统PPT课件.pptx
- 智慧路边停车管理系统方案.pptx
- 【企业内控系列】企业内部控制之人力资源管理控制(17页).doc
- 温度传感器分类与特点.pptx
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功
评论0