Acceso al sitio web del Consejo de Políticas de Infraestructura

   
ARTÍCULO
TITULO

Object boundaries inconsistencies detection method for 2D-3D conversion results and depth maps

Stanislav Dolganov    
Dmitriy Vatolin    

Resumen

The creation of S3D movies by converting 2D captured footage often introduces depth-map inaccuracies. Such artifacts can significantly degrade the viewing experience even if they occur only in unsalient background objects. In this paper we propose a method for detecting foreground objects that are stuck to the background. Our method extracts information about motion in the scene and detects conversion-related discrepancies between motion strength and depth. We demonstrate the performance of the method by applying it to 39 full-length converted 3D movies and by providing the results of our analysis as well as examples of detected problem shots.

PÁGINAS
pp. 1 - 8

 Artículos similares

       
 
Yanhong Lin, Faming Huang, Yang Zhang, Qi Wang, Jinliang Huang, Qiuming Chen and Xiongzhi Xue    
Planning and zoning of marine areas is a common practice worldwide. In China, marine major function-oriented zoning (MMFOZ) is an effective geographical method for ensuring sustainable marine resource and ecosystems development; however, ecosystem-based ... ver más

 
Vladimir Skripachev,Mikhail Guida,Nikolay Guida,Alexander Zhukov     Pág. 53 - 61
The article discusses in detail the use of collapsible neural networks (CNNs) for use in solving problems of detecting objects in aerospace images. The structure of collapsing neural networks is revealed, taking into account the specifics of the applicat... ver más

 
Fridolin Wild, Lawrence Marshall, Jay Bernard, Eric White and John Twycross    
The integration of augmented reality (AR) technology into personal computing is happening fast, and augmented workplaces for professionals in areas such as Industry 4.0 or digital health can reasonably be expected to form liminal zones that push the boun... ver más
Revista: Information

 
Guangqi Liu, Qinghai Ding, Haibo Luo, Moran Ju, Tianming Jin, Miao He and Gang Dong    
Development of an accurate and automated algorithm to completely segment cervical cells in Pap images is still one of the most challenging tasks. The main reasons are the presence of overlapping cells and the lack of guiding mechanism for the convergence... ver más
Revista: Applied Sciences

 
Benjamin Saß, Barbara Carl, Mirza Pojskic, Christopher Nimsky and Miriam Bopp    
Background: Implementation of intraoperative 3D ultrasound (i3D US) into modern neuronavigational systems offers the possibility of live imaging and subsequent imaging updates. However, different modalities, image acquisition strategies, and timing of im... ver más
Revista: Applied Sciences