Accuracy analysis of surface normal reconstruction in stereo vision

Harms, Hans-Heinrich; Beck, Johannes; Ziegler, Julius; Stiller, Christoph

doi:10.1109/ivs.2014.6856436

Cited by 10 publications

(5 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The challenge is, however, to fit such a model to the incomplete and deformed point cloud captured with a depth camera. On the one hand, simple least square methods [41] fail due to outliers, and methods using surface normals [42] cannot be applied due to noise from pleats on the clothes. On the other hand, complex object fitting approaches [43] are unsuitable, since the high processing time prevents any real-time evaluation.…”

Section: Approximation Of Shank Axismentioning

confidence: 99%

Real-time limb tracking in single depth images based on circle matching and line fitting

et al. 2021

View full text Add to dashboard Cite

Modern lower limb prostheses neither measure nor incorporate healthy residual leg information for intent recognition or device control. In order to increase robustness and reduce misclassification of devices like these, we propose a vision-based solution for real-time 3D human contralateral limb tracking (CoLiTrack). An inertial measurement unit and a depth camera are placed on the side of the prosthesis. The system is capable of estimating the shank axis of the healthy leg. Initially, the 3D input is transformed into a stabilized coordinate system. By splitting the subsequent shank estimation problem into two less computationally intensive steps, the computation time is significantly reduced: First, an iterative closest point algorithm is applied to fit circular models against 2D projections. Second, the random sample consensus method is used to determine the final shank axis. In our study, three experiments were conducted to validate the static, the dynamic and the real-world performance of our CoLiTrack approach. The shank angle can be tracked at 20 Hz for one sixth of the entire human gait cycle with an angle estimation error below $$2.8\pm 2.1^{\circ }$$ 2.8 ± 2 . 1 ∘ . Our promising results demonstrate the robustness of the novel CoLiTrack approach to make “next-generation prostheses” more user-friendly, functional and safe.

show abstract

Section: Approximation Of Shank Axismentioning

confidence: 99%

Real-time limb tracking in single depth images based on circle matching and line fitting

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Yet, it is possible to extract normal surface vectors from depth data. [8] [26] [6] [21] proposes different approaches to estimate surface normal vectors from disparity images. Those methods calculate the depth values from disparity data and then estimate normal surface vectors.…”

Section: Surface Featuresmentioning

confidence: 99%

Environment reconstruction on depth images using Generative Adversarial Networks

Matias¹,

Souza²,

Wolf³

2019

Preprint

View full text Add to dashboard Cite

Robust perception systems are essential for autonomous vehicle safety. To navigate in a complex urban environment, it is necessary precise sensors with reliable data. The task of understanding the surroundings is hard by itself; for intelligent vehicles, it is even more critical due to the high speed in which the vehicle navigates. To successfully navigate in an urban environment, the perception system must quickly receive, process, and execute an action to guarantee both passenger and pedestrian safety. Stereo cameras collect environment information at many levels, e.g., depth, color, texture, shape, which guarantee ample knowledge about the surroundings. Even so, when compared to human, computational methods lack the ability to deal with missing information, i.e., occlusions. For many perception tasks, this lack of data can be a hindrance due to the environment incomplete information. In this paper, we address this problem and discuss recent methods to deal with occluded areas inference. We then introduce a loss function focused on disparity and environment depth data reconstruction, and a Generative Adversarial Network (GAN) architecture able to deal with occluded information inference. Our results present a coherent reconstruction on depth maps, estimating regions occluded by different obstacles. Our final contri-

show abstract

“…Some of these factors have been studied, see e.g. (Aguilar et al, 1996;Harms et al, 2014) and references therein. In order to verify if photogrammetric and other camera properties affect the robustness, we employ image sets taken with different cameras.…”

Section: Automated Isb Reconstructionmentioning

confidence: 99%