Combining Edge Images and Depth Maps for Robust Visual Odometry

Schenk, Fabian; Fraundorfer, Friedrich

doi:10.5244/c.31.149

Cited by 9 publications

(8 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[21] and our visual odometer Pl-EVO. In the table, column REVO corresponds to the algorithm available online, which only register two consecutive frames, whereas column REVO E+D+Opt corresponds to the algorithm described in [13], where they optimize for the N previous frames to improve camera motion estimation. Table I suggests that the Pl-EVO obtains better results than the state-of-the-art odometers and SLAM systems for indoor environments lacking texture, without the need for relocalization or loop closing strategies and only performing a register between two keyframes.…”

Section: Resultsmentioning

confidence: 99%

“…† The results come from [21]. ‡ The results come from [13] * It shows a possible loop closing dataset.…”

Section: Resultsmentioning

confidence: 99%

“…Using these features individually have some drawbacks, while combining them with other features tend to improve the registration performance. Some combined methods use lines and keypoints [4] [5], lines, keypoints and planes [6] [7], keypoints and planes [8] [9], semi-direct methods and planes [10], direct methods and edges [11], points with planar patches [12] and edges with depth maps [13].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Combination of Planes and Image Edges for a Visual Odometer for Weakly Textured Structured Environments

Company-Corcoles¹,

Ortiz²

2019

2019 24th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)

View full text Add to dashboard Cite

Camera trajectory estimation has generated a lot of interest during the last decades, especially for robotic positioning. It is well-known that outdoors positioning mainly relies on GPS, whereas one of the main used methods in indoor positioning is visual odometry. As it is well known, visual odometers get typically in trouble when the environment is weakly textured. Facing this situation, this paper develops and tests a novel visual odometer that combines image edges and planar information to estimate the trajectory of an RGB-D camera in environments that lack texture. We also present a plane matching method based on a graph matching technique. To conclude, a comparison of the proposed odometer for two well-known datasets and other visual odometers and SLAM systems is reported. The comparison shows our method as more accurate as for the estimation of the position in indoor places where visual features are poor, while similar values are obtained for other indoor environments.

show abstract

Section: Resultsmentioning

confidence: 99%

“…† The results come from [21]. ‡ The results come from [13] * It shows a possible loop closing dataset.…”

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Combination of Planes and Image Edges for a Visual Odometer for Weakly Textured Structured Environments

Company-Corcoles¹,

Ortiz²

2019

2019 24th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)

View full text Add to dashboard Cite

show abstract

“…In [64], direct edge alignment is proposed that minimizes the sum of squared distances between the reprojected and the nearest edge point using the distance transform of the edge-map. Other works propose to jointly minimize this edge distance and other errors, e.g., a photometric error [114] and an ICP-based point-to-plane distance [93]. Later works such as [124] and [56] take the image gradient direction also into account for the direct edge alignment.…”

Section: Edge-based Methodsmentioning

confidence: 99%

RGB-D Odometry and SLAM

Civera

Lee

2019

Advances in Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

The emergence of modern RGB-D sensors had a significant impact in many application fields, including robotics, augmented reality (AR) and 3D scanning. They are low-cost, low-power and low-size alternatives to traditional range sensors such as LiDAR. Moreover, unlike RGB cameras, RGB-D sensors provide the additional depth information that removes the need of frame-by-frame triangulation for 3D scene reconstruction. These merits have made them very popular in mobile robotics and AR, where it is of great interest to estimate egomotion and 3D scene structure. Such spatial understanding can enable robots to navigate autonomously without collisions and allow users to insert virtual entities consistent with the image stream. In this chapter, we review common formulations of odometry and Simultaneous Localization and Mapping (known by its acronym SLAM) using RGB-D stream input. The two topics are closely related, as the former aims to track the incremental camera motion with respect to a local map of the scene, and the latter to jointly estimate the camera trajectory and the global map with consistency. In both cases, the standard approaches minimize a cost function using nonlinear optimization techniques. This chapter consists of three main parts: In the first part, we introduce the basic concept of odometry and SLAM and motivate the use of RGB-D sensors. We also give mathematical preliminaries relevant to most odometry and SLAM algorithms. In the second part, we detail the three main components of SLAM systems: camera pose tracking, scene mapping and loop closing. For each component, we describe different approaches proposed in the literature. In the final part, we provide a brief discussion on advanced research topics with the references to the state-of-the-art.

show abstract

“…To achieve the vision-based localization against the model, the authors propose an edge alignment scheme for the current image against a virtual image extracted from the model that is used as a reference or keyframe. A visual odometry framework (Schenk & Fraundorfer, 2017) using a similar algorithm for RGB-D sensors provides a better evaluation on the approach and produces estimates with a drift accumulation on par with state-of-the-art visual odometry (VO) methods.…”

Section: State Of the Artmentioning

confidence: 99%

Overview obstacle maps for obstacle‐aware navigation of autonomous drones

Pestana

Maurer

Muschick

et al. 2019

Journal of Field Robotics

Self Cite

View full text Add to dashboard Cite

Achieving the autonomous deployment of aerial robots in unknown outdoor environments using only onboard computation is a challenging task. In this study, we have developed a solution to demonstrate the feasibility of autonomously deploying drones in unknown outdoor environments, with the main capability of providing an obstacle map of the area of interest in a short period of time. We focus on use cases where no obstacle maps are available beforehand, for instance, in search and rescue scenarios, and on increasing the autonomy of drones in such situations. Our vision‐based mapping approach consists of two separate steps. First, the drone performs an overview flight at a safe altitude acquiring overlapping nadir images, while creating a high‐quality sparse map of the environment by using a state‐of‐the‐art photogrammetry method. Second, this map is georeferenced, densified by fitting a mesh model and converted into an Octomap obstacle map, which can be continuously updated while performing a task of interest near the ground or in the vicinity of objects. The generation of the overview obstacle map is performed in almost real time on the onboard computer of the drone, a map of size 100m×75m is created in ≈2.75min, therefore, with enough time remaining for the drone to execute other tasks inside the area of interest during the same flight. We evaluate quantitatively the accuracy of the acquired map and the characteristics of the planned trajectories. We further demonstrate experimentally the safe navigation of the drone in an area mapped with our proposed approach.

show abstract

Combining Edge Images and Depth Maps for Robust Visual Odometry

Cited by 9 publications

References 12 publications

Combination of Planes and Image Edges for a Visual Odometer for Weakly Textured Structured Environments

Combination of Planes and Image Edges for a Visual Odometer for Weakly Textured Structured Environments

RGB-D Odometry and SLAM

Overview obstacle maps for obstacle‐aware navigation of autonomous drones

Contact Info

Product

Resources

About