Multi-Kinect scene reconstruction: Calibration and depth inconsistencies

Yang, Roy Sirui; Chan, Yuk Hin; Gong, Rui; Nguyen, Minh; Gastélum-Strozzi, Alfonso; Delmas, Patrice; Gimel’farb, Georgy; Ababou, Rachel

doi:10.1109/ivcnz.2013.6726991

Cited by 27 publications

(25 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Early solutions addressing multi-view reconstruction by Kinect sensors suffered from severe drawbacks derived from the limitations of the available sensing hardware. A multi-Kinect system to seamlessly render a scene from a wide range of angles was described in [6]. The main drawback of related solutions relates to depth inconsistencies between the measurements acquired with different Kinect devices.…”

Section: Related Workmentioning

confidence: 99%

Multiview 3D sensing and analysis for high quality point cloud reconstruction

Satnik

Izquierdo

Orjesek

2018

Tenth International Conference on Machine Vision (ICMV 2017)

View full text Add to dashboard Cite

Multiview 3D reconstruction techniques enable digital reconstruction of 3D objects from the real world by fusing different viewpoints of the same object into a single 3D representation. This process is by no means trivial and the acquisition of high quality point cloud representations of dynamic 3D objects is still an open problem. In this paper, an approach for high fidelity 3D point cloud generation using low cost 3D sensing hardware is presented. The proposed approach runs in an efficient low-cost hardware setting based on several Kinect v2 scanners connected to a single PC. It performs autocalibration and runs in real-time exploiting an efficient composition of several filtering methods including Radius Outlier Removal (ROR), Weighted Median filter (WM) and Weighted Inter-Frame Average filtering (WIFA). The performance of the proposed method has been demonstrated through efficient acquisition of dense 3D point clouds of moving objects.

show abstract

Section: Related Workmentioning

confidence: 99%

Multiview 3D sensing and analysis for high quality point cloud reconstruction

Satnik

Izquierdo

Orjesek

2018

Tenth International Conference on Machine Vision (ICMV 2017)

View full text Add to dashboard Cite

show abstract

“…Since Microsoft released the SDK for the Kinect, combining multiple Kinects to improve accuracy has become feasible. In [53], a system with multiple Kinects is built to reconstruct an object in 3D coordinates, and the uncertainties of the Kinect's depth sensor and a robust way to properly align multiple Kinects are also analyzed and proposed.…”

Section: Background and Motivationmentioning

confidence: 99%

“…Currently, a matrix barcode on a planar surface, i.e., quick response codes (QR Codes), is the most wildly used technology in the optical identification field. Open-source libraries for QR codes has been released by OpenCV and can be applied in various types of optical cameras such as Optitrack [36], Kinect [53], and Wii remotes [38]. In addition, calibration objects can be designed with special shapes to be easily recognizable by depth sensors [23] [19].…”

Section: Calibrationmentioning

confidence: 99%

Development of a Self-Calibrated Motion Capture System by Nonlinear Trilateration of Multiple Kinects v2

2017

View full text Add to dashboard Cite

In this paper, a Kinect-based distributed and real-time motion capture system is developed. A trigonometric method is applied to calculate the relative positions of Kinect v2 sensors with a calibration wand and register the sensors' positions automatically. By combining results from multiple sensors with a nonlinear least square method, the accuracy of motion capture is optimized. Moreover, to exclude inaccurate results from sensors, a computational geometry is applied in the occlusion approach to discover occluded joint data. The synchronization approach is based on the NTP protocol, which synchronizes the time between the clocks of a server and of clients dynamically, leading to the proposed system being real time. Experiments to validate the proposed system are conducted from the perspective of calibration, occlusion, and accuracy. More specifically, the mean absolute error of the calibration results is 0.73 cm, the proposed occlusion method is tested on upper and lower limbs, and the synchronization component guarantees the clock synchronization and real-time performance for more than 99% of the measurement process.Furthermore, to demonstrate the practical performance of our system, a comparison with previously developed motion capture systems (the linear trilateration approach [52] and the geometric trilateration approach [51]) with the benchmark OptiTrack system is performed for the tracked joints of the head, shoulder, elbow, and wrist, therein showing that the accuracy of our proposed system is 38.3% and 24.1% better than the aforementioned two trilateration systems. Quantitative analysis is also conducted on our proposed system with the commercial inertial motion capture system Delsys smart sensor system by comparing the measurements of lower limbs (i.e., hips, knees, and ankles), and the standard deviation of our proposed system's measurement results is 4.92 cm.

show abstract

“…A possible solution is to use an infrared depth sensor integrated to a camera, allowing a 3D reconstruction of the scenario. Kinect and Asus Xtion Pro are examples of systems that provide VGA resolution 3D measurement with high accuracy and affordable prices [6]. An intrinsic problem to this solution is that infrared sensors do not work well outdoors, turning it difficult to use these devices for activities such as can collecting at a beach [7].…”

Section: Computer Visionmentioning

confidence: 99%

Modular robot used as a beach cleaner

Roza

Silva

Pereira

et al. 2016

Ingeniare. Rev. chil. ing.

View full text Add to dashboard Cite

This paper presents the development of a modular mobile robot platform for multiple purposes and its adaptation into a beach cleaner setup. The robot has a robust construction allowing it to endure several sorts of environments. In the presented configuration, the developed robot is able to autonomously collect cans using an excavator-like claw. In addition to detailing the system's design and construction, this paper presents the description of the developed embedded electronics modules, a motor closed-loop speed control system and the optical flow algorithms that allows the computer vision system to detect and avoid obstacles and track the cans to be collected.Keywords: Mobile robots, robotics, computer vision, optical flow, speed control. RESUMEN

show abstract

Multi-Kinect scene reconstruction: Calibration and depth inconsistencies

Cited by 27 publications

References 10 publications

Multiview 3D sensing and analysis for high quality point cloud reconstruction

Multiview 3D sensing and analysis for high quality point cloud reconstruction

Development of a Self-Calibrated Motion Capture System by Nonlinear Trilateration of Multiple Kinects v2

Modular robot used as a beach cleaner

Contact Info

Product

Resources

About