Multi-Frame Based Homography Estimation for Video Stitching in Static Camera Environments

Park, Keon-woo; Shim, Yoo-Jeong; Lee, Myeong-jin; Ahn, Heejune

doi:10.3390/s20010092

Cited by 11 publications

(12 citation statements)

References 32 publications

(58 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This makes training the network slightly harder, additionally it may somewhat reduce the quality of the results due to network somewhat trying to adjust to observation angle, therefore making the IoU metric values lower, despite visually being feasible. Solving the perspective invariance may also be a partial solution to the homography [67,68] problem as our reconstructed object would already be rotated with respect to the camera space.…”

Section: Discussionmentioning

confidence: 99%

3D Object Reconstruction from Imperfect Depth Data Using Extended YOLOv3 Network

Kulikajevas

Maskeliūnas

Damaševičius

et al. 2020

Sensors

View full text Add to dashboard Cite

State-of-the-art intelligent versatile applications provoke the usage of full 3D, depth-based streams, especially in the scenarios of intelligent remote control and communications, where virtual and augmented reality will soon become outdated and are forecasted to be replaced by point cloud streams providing explorable 3D environments of communication and industrial data. One of the most novel approaches employed in modern object reconstruction methods is to use a priori knowledge of the objects that are being reconstructed. Our approach is different as we strive to reconstruct a 3D object within much more difficult scenarios of limited data availability. Data stream is often limited by insufficient depth camera coverage and, as a result, the objects are occluded and data is lost. Our proposed hybrid artificial neural network modifications have improved the reconstruction results by 8.53% which allows us for much more precise filling of occluded object sides and reduction of noise during the process. Furthermore, the addition of object segmentation masks and the individual object instance classification is a leap forward towards a general-purpose scene reconstruction as opposed to a single object reconstruction task due to the ability to mask out overlapping object instances and using only masked object area in the reconstruction process.

show abstract

Section: Discussionmentioning

confidence: 99%

3D Object Reconstruction from Imperfect Depth Data Using Extended YOLOv3 Network

Kulikajevas

Maskeliūnas

Damaševičius

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…an immersive 2D screen, a VR capable smartphone, and a VR headset. The VR contents are created by stitching video streams [28,29,30] from multiple cameras attached to a rig. We use two different methods to improve the geometric alignment [31,32] of stitched video streams i.e.…”

Section: Presence Quality and Cybersickness In Virtual Realitymentioning

confidence: 99%

A study on presence quality and cybersickness in 2D, smartphone, and VR

2022

KSII TIIS

View full text Add to dashboard Cite

Recent improvements in technology have increased the consumption of virtual reality (VR) contents on immersive displays. The VR experience depends on the type of displays as well as the quality of VR contents. However, research on the impacts of VR content quality on VR experience and comparisons among different types of immersive display devices are lacking. In this study, VR contents created with our VR framework, are provided to participants on conventional two-dimensional (2D) immersive displays and VR headset. The geometric alignment of VR contents is improved with the addition of two calibration modes (i.e. preprocessing and straightening). The subjective feelings of presence and cybersickness experienced by participants while consuming VR contents created by our framework and commercial solutions are recorded in the form of questionnaires. The results of this study indicate that the improvements in VR quality lead to a better presence and less cybersickness in both conventional 2D displays and VR headset. Furthermore, the level of presence and cybersickness increases in VR headsets as compared to conventional 2D displays. Finally, the VR content quality improvements lead to a better VR experience for our VR framework as compared to commercial solutions.

show abstract

“…However, these methods hide rather than reduce the misalignment distortion and thus are not a fundamental solution to reduce distortion. In our previous works, to reduce the misalignment distortion for scenes with plain textures or few feature points, we estimated homography using feature points extracted during a predetermined time interval [21,22]. We showed experimentally that these methods could reduce misalignment distortion in overlapped regions in video sequences without foreground objects [21,22].…”

Section: Seam-based Stitchingmentioning

confidence: 99%

Region-Based Static Video Stitching for Reduction of Parallax Distortion

Park

Shim

Lee

2021

Sensors

Self Cite

View full text Add to dashboard Cite

In this paper, we propose a semantic segmentation-based static video stitching method to reduce parallax and misalignment distortion for sports stadium scenes with dynamic foreground objects. First, video frame pairs for stitching are divided into segments of different classes through semantic segmentation. Region-based stitching is performed on matched segment pairs, assuming that segments of the same semantic class are on the same plane. Second, to prevent degradation of the stitching quality of plain or noisy videos, the homography for each matched segment pair is estimated using the temporally consistent feature points. Finally, the stitched video frame is synthesized by stacking the stitched matched segment pairs and the foreground segments to the reference frame plane by descending order of the area. The performance of the proposed method is evaluated by comparing the subjective quality, geometric distortion, and pixel distortion of video sequences stitched using the proposed and conventional methods. The proposed method is shown to reduce parallax and misalignment distortion in segments with plain texture or large parallax, and significantly improve geometric distortion and pixel distortion compared to conventional methods.

show abstract

Multi-Frame Based Homography Estimation for Video Stitching in Static Camera Environments

Cited by 11 publications

References 32 publications

3D Object Reconstruction from Imperfect Depth Data Using Extended YOLOv3 Network

3D Object Reconstruction from Imperfect Depth Data Using Extended YOLOv3 Network

A study on presence quality and cybersickness in 2D, smartphone, and VR

Region-Based Static Video Stitching for Reduction of Parallax Distortion

Contact Info

Product

Resources

About