Very High Frame Rate Volumetric Integration of Depth Images on Mobile Devices

Kähler, Olaf; Prisacariu, Victor Adrian; Ren, Carl Yuheng; Sun, Xin; Torr, Philip H. S.; Murray, David

doi:10.1109/tvcg.2015.2459891

Cited by 266 publications

(246 citation statements)

References 21 publications

Supporting

Mentioning

246

Contrasting

Order By: Relevance

“…It should also be noted that pose tracking 3D model generation from depth camera frames is an active area of research and we expect future depth cameras to exhibit improved performance for both aspects of operation. 37,38 Fabrication errors in the compensator would also be expected to improve with additional advancements in molding technique. The 3D-printed compensator shapes were evaluated to be within 0.1 mm of the desired thickness, indicating that errors were introduced in the molding process.…”

Section: Discussionmentioning

confidence: 99%

A depth‐sensing technique on 3D‐printed compensator for total body irradiation patient measurement and treatment planning

et al. 2016

View full text Add to dashboard Cite

Purpose: The purpose of total body irradiation (TBI) techniques is to deliver a uniform radiation dose to the entire volume of a patient's body. Due to variations in the thickness of the patient, it is difficult to produce such a uniform dose distribution throughout the body. In many techniques, a compensator is used to adjust the dose delivered to various sections of the patient. The current study aims to develop and validate an innovative method of using depth-sensing cameras and 3D printing techniques for TBI treatment planning and compensator fabrication. Methods: A tablet with an integrated depth-sensing camera and motion tracking sensors was used to scan a RANDO™ phantom positioned in a TBI treatment booth to detect and store the 3D surface in a point cloud format. The accuracy of the detected surface was evaluated by comparing extracted body thickness measurements with corresponding measurements from computed tomography (CT) scan images. The thickness, source to surface distance, and off-axis distance of the phantom at different body section were measured for TBI treatment planning. A detailed compensator design was calculated to achieve a uniform dose distribution throughout the phantom. The compensator was fabricated using a 3D printer, silicone molding, and a mixture of wax and tungsten powder. In vivo dosimetry measurements were performed using optically stimulated luminescent detectors. Results: The scan of the phantom took approximately 30 s. The mean error for thickness measurements at each section of phantom relative to CT was 0.48 ± 0.27 cm. The average fabrication error for the 3D-printed compensator was 0.16 ± 0.15 mm. In vivo measurements for an end-to-end test showed that overall dose differences were within 5%. Conclusions: A technique for planning and fabricating a compensator for TBI treatment using a depth camera equipped tablet and a 3D printer was demonstrated to be sufficiently accurate to be considered for further investigation. C

show abstract

Section: Discussionmentioning

confidence: 99%

A depth‐sensing technique on 3D‐printed compensator for total body irradiation patient measurement and treatment planning

et al. 2016

View full text Add to dashboard Cite

show abstract

“…For the depth map fusion we use the approach presented by (Kähler et al, 2015). This approach is highly optimized for real time processing on a GPU.…”

Section: Depth Map Fusion With Semantic Filteringmentioning

confidence: 99%

“…We exploit the hashing of (Nießner et al, 2013) to be unconstrained in size of the scene. To be able to use our aligned depth maps, we extended the framework of (Kähler et al, 2015) to take camera poses as input for the depth maps positioning and not to perform any tracking.…”

Section: Depth Map Fusion With Semantic Filteringmentioning

confidence: 99%

“…Whereas this approach is well suited for man-made structures, it is clearly visible that the approach is not eligible for densification of vegetation due to the high structure and nonplanarity of vegetation. The same scene has been reconstructed using our dense reconstruction approach based on the DiscreteContinuous Dense Image Matching of (Shekhovtsov et al, 2016) and depth map fusion of (Kähler et al, 2015). We gain well reconstructed vegetation (see Figure 13 right) suitable for further inspection tasks.…”

Section: Reconstruct-ability Of Vegetationmentioning

confidence: 99%

See 1 more Smart Citation

Automated Inspection of Power Line Corridors to Measure Vegetation Undercut Using Uav-Based Images

Maurer

Höfer²,

Fraundorfer

et al. 2017

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:Power line corridor inspection is a time consuming task that is performed mostly manually. As the development of UAVs made huge progress in recent years, and photogrammetric computer vision systems became well established, it is time to further automate inspection tasks. In this paper we present an automated processing pipeline to inspect vegetation undercuts of power line corridors. For this, the area of inspection is reconstructed, geo-referenced, semantically segmented and inter class distance measurements are calculated. The presented pipeline performs an automated selection of the proper 3D reconstruction method for on the one hand wiry (power line), and on the other hand solid objects (surrounding). The automated selection is realized by performing pixel-wise semantic segmentation of the input images using a Fully Convolutional Neural Network. Due to the geo-referenced semantic 3D reconstructions a documentation of areas where maintenance work has to be performed is inherently included in the distance measurements and can be extracted easily. We evaluate the influence of the semantic segmentation according to the 3D reconstruction and show that the automated semantic separation in wiry and dense objects of the 3D reconstruction routine improves the quality of the vegetation undercut inspection. We show the generalization of the semantic segmentation to datasets acquired using different acquisition routines and to varied seasons in time.

show abstract

“…One of the main scientific and technological achievements at the start of this trend is undoubtedly Kinect Fusion, by Newcombe et al [1], shortly followed by several extensions [2] [3] and alternative formulations of the original problem and solution [4] [5]. At the core of these algorithms is an elegant method for volumetric integration of depth information into a truncated signed distance field (TSDF).…”

Section: Introductionmentioning

confidence: 99%

From Feature Detection in Truncated Signed Distance Fields to Sparse Stable Scene Graphs

Canelhas

Stoyanov

Lilienthal

2016

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Abstract-With the increased availability of GPUs and multicore CPUs, volumetric map representations are an increasingly viable option for robotic applications. A particularly important representation is the truncated signed distance field (TSDF) that is at the core of recent advances in dense 3D mapping. However, there is relatively little literature exploring the characteristics of 3D feature detection in volumetric representations. In this paper we evaluate the performance of features extracted directly from a 3D TSDF representation. We compare the repeatability of Integral invariant features, specifically designed for volumetric images, to the 3D extensions of Harris and Shi & Tomasi corners. We also study the impact of different methods for obtaining gradients for their computation. We motivate our study with an example application for building sparse stable scene graphs, and present an efficient GPU-parallel algorithm to obtain the graphs, made possible by the combination of TSDF and 3D feature points. Our findings show that while the 3D extensions of 2D corner-detection perform as expected, integral invariants have shortcomings when applied to discrete TSDFs. We conclude with a discussion of the cause for these points of failure that sheds light on possible mitigation strategies.

show abstract

Very High Frame Rate Volumetric Integration of Depth Images on Mobile Devices

Cited by 266 publications

References 21 publications

A depth‐sensing technique on 3D‐printed compensator for total body irradiation patient measurement and treatment planning

A depth‐sensing technique on 3D‐printed compensator for total body irradiation patient measurement and treatment planning

Automated Inspection of Power Line Corridors to Measure Vegetation Undercut Using Uav-Based Images

From Feature Detection in Truncated Signed Distance Fields to Sparse Stable Scene Graphs

Contact Info

Product

Resources

About