Weight-flooding aggregation with canny edge constraint for 3D depth estimation

Zi Shiung Tsai, Pau Choo Chung, Kuan Wei Chi

研究成果: Article

摘要

Considering the computation efficiency, local based 3D depth estimation has been considered of higher potential compared with global based m ethods. The most well recognized local based 3D depth estim ation is the cross-based approach, which computes a local region for each pixel, called the support region, and the cost aggregation for the pixel is conducted within the support region. However, due to the optim ization is conducted only within the support region, it also has less accuracy performance. Furthermore, obtaining the support region requires a significant computing power. In view of these pr oblems, this paper proposes a novel local based m ethod using weight-flooding aggregation for cost optimization in 3D depth estim ation. The weightflooding aggregation performs in a global way, such that every pixel transfers its cost to its neighboring pixels through a weight map. Accompanied with the weight-flooding aggregation is an edge based restriction to reduce the cost influences fr om other objects during the propagation. Any propagation passing through the edge will be stopped. The propagation in the flooding is per formed from a gradual collection. Thus part of the flooding propagation to a pixel can be r eused by the neighboring pixels and an integr al computing can be applied. By so designed, the pixel optim ization is allowed to receive costs globally, achieving high accuracy. Also, the lift of the requirem ent of obtaining support region for each pixel and the feasibility of applying integral computing significantly reduces the time complexity. In this paper, the L-R check is adopted as a post pr ocessing to rectify the errors which may result from mismatch due to occlusion or noise. The experimental results show that compared to cross-based local method, the system requires only half to onethird computing cost while achieves comparable estimation accuracy.

原文English
頁(從 - 到)1495-1519
頁數25
期刊Journal of Information Science and Engineering
31
發行號5
出版狀態Published - 2015 九月 1

    指紋

All Science Journal Classification (ASJC) codes

  • Software
  • Human-Computer Interaction
  • Hardware and Architecture
  • Library and Information Sciences
  • Computational Theory and Mathematics

引用此