DSO(Direct Sparse Odometry),是慕尼黑工业大学(Technical University of Munich, TUM)计算机视觉实验室的雅各布.恩格尔(Jakob Engel)博士,于2016年发布的一个视觉里程计方法,在SLAM领域,DSO属于稀疏直接法,它不是完整的SLAM,因为它不包含回环检测、地图复用的功能。因此,它不可避免地会出现累计误差,尽管很小,但不能消除。 虽然代码开源,但是正如高翔博士所说“由于某些历史和个人的原因,DSO的代码清晰度和可读性,明显弱于其他SLAM方案如ORB、SVO、okvis等,使得研究人员很难以它为基础,展开后续的研究工作。”网上DSO相关的资料还是非常少,甚至少得可怜,这里也只能跑个demo,测试一下,有兴趣的可以尝试修改一下。 项目主页: https://vision.in./research/vslam/dso 内容摘自高翔博士知乎博客: https://zhuanlan.zhihu.com/p/29177540 内容介绍 DSO是少数使用纯直接法(Fully direct)计算视觉里程计的系统之一。相比SVO,ORB-SLAM2等方案,从方法上来说,DSO是新颖、独树一帜的。 直接法相比于特征点法,有两个非常不同的地方:
DSO直接法里程计以更整体、更优雅的方式处理了数据关联问题。特征点法需要依赖重复性较强的特征提取器,以及正确的特征匹配,才能得正确地计算相机运动。在环境纹理较好,角点较多时,这当然是可行的——不过直接法在这种环境下也能正常工作。然而,如果环境中出现:(1)环境中存在许多重复纹理; (2)环境中缺乏角点,出现许多边缘或光线变量不明显区域。 DSO也将运行困难。 英文介绍 DSO is a novel direct and sparse formulation for Visual Odometry. It combines a fully direct probabilistic model (minimizing a photometric error) with consistent, joint optimization of all model parameters, including geometry - represented as inverse depth in a reference frame - and camera motion. This is achieved in real time by omitting the smoothness prior used in other direct methods and instead sampling pixels evenly throughout the images. DSO does not depend on keypoint detectors or descriptors, thus it can naturally sample pixels from across all image regions that have intensity gradient, including edges or smooth intensity variations on mostly white walls. The proposed model integrates a full photometric calibration, accounting for exposure time, lens vignetting, and non-linear response functions. We thoroughly evaluate our method on three different datasets comprising several hours of video. The experiments show that the presented approach significantly outperforms state-of-the-art direct and indirect methods in a variety of real-world settings, both in terms of tracking accuracy and robustness. 参考文献 高博知乎: https://zhuanlan.zhihu.com/p/29177540 资源 三维点云论文及相关应用分享 【点云论文速读】基于激光雷达的里程计及3D点云地图中的定位方法 3D-MiniNet: 从点云中学习2D表示以实现快速有效的3D LIDAR语义分割(2020) PCL中outofcore模块---基于核外八叉树的大规模点云的显示 更多文章可查看:点云学习历史文章大汇总 SLAM及AR相关分享 |
|