DJI在CVPR15上发的关于GUIDANCE组件的文章
入手了GUIDANCE一周,这里对这篇文章细细品味一下。
由于缺乏一个强大的视觉感知平台,开发了一个易于扩展的数据采集平台。
这个论文可以主要参考一下DJI人员开发GUIDANCE时候,内置了什么样的算法。
1.Introduction
提到了VO(Visual Odometry)的概念,SLAM是VO[1]的扩展。关于SLAM[2][3],是用于机器人定位位置识别。大部分是单目的,这里也有双目的方法[4]。这里GUIDANCE包括了5个方向的双目,可以提供深度图,超声数据,障碍物距离等内容。接下来,故事的展开分为内置函数,SDK接口以及相关应用
2.内置:Guidance是升级版的[6]
4.可扩展的智能应用:
1.Autonomous Navigation
拥有高精度的GPS定位
2.Visual SLAM[8]
依赖高精度的VO结果
3.Depth Based Tracking
为了开发SLAM避障,请阅读下面黑体
[1]C. Forster, M. Pizzoli, and D. Scaramuzza. SVO: Fast semi- direct monocular visual odometry. In ICRA, 2014.
[2]J. Engel, T. Schops, and D. Cremers. LSD-SLAM: Large- scale direct monocular SLAM. In ECCV, 2014.
[3]A. Davison. Real-time simultaneous localisation and map- ping with a single camera. In Computer Vision, 2003. Proceedings. Ninth IEEE International Conference on, pages 1403–1410 vol.2, Oct 2003
[4] F. Endres, J. Hess, N. Engelhard, J. Sturm, D. Cremers, and W. Burgard. An evaluation of the RGB-D SLAM system. In Robotics and Automation (ICRA), 2012 IEEE International Conference on, pages 1691–1696, May 2012.
[5]D. Honegger, L. Meier, P. Tanskanen, and M. Pollefeys. An open source and open hardware embedded metric optical flow cmos camera for indoor and outdoor applications. In Robotics and Automation (ICRA), IEEE International Con- ference on, pages 1736–1741, May 2013.
[6]G. Zhou, A. Liu, K. Yang, T. Wang, and Z. Li. An embedded solution to visual mapping for consumer drones. In Computer Vision and Pattern Recognition Workshops (CVPRW), IEEE Conference on, pages 670–675, 2014.
[7]B. D. Lucas and T. Kanade. An iterative image registration technique with an application to stereo vision. In Proceed- ings of the 7th International Joint Conference on Artificial
Intelligence - Volume 2, pages 674–679, 1981
[8]G. Klein and D. Murray. Parallel tracking and mapping for small AR workspaces. In Sixth IEEE and ACM Internation- al Symposium on Mixed and Augmented Reality (ISMAR), 2007.