A Fast TGV-l1 RGB-D Flow Estimation

Title
A Fast TGV-l1 RGB-D Flow Estimation
Authors
노준하임화섭안상철
Keywords
RGBD Flow; Scene Flow; TGV-l1
Issue Date
2014-12
Publisher
Lecture notes in computer science
Citation
VOL 8887, 151-161
Abstract
We present a novel method for fast and dense 3D scene flow estimation which optimizes consistency and smoothness in both intensity and depth data while considering computing e ciency for the real-world applications. 3D scene flow estimation is an attractive problem with the advent of commodity RGB-D cameras. Naive extensions of recent variational optical flow techniques show promising but limited successes. Due to their primitive priors, solutions from total variation approaches prefer unrealistic constant motion. To overcome these problems and consider the computational e ciency, we adopt an image-guided total generalized variation (ITGV) regularization. As demonstrated with experimental results, the proposed method outperforms both in terms of accuracy and speed compared to the existing variational approaches.
URI
http://pubs.kist.re.kr/handle/201004/49018
ISSN
03029743
Appears in Collections:
KIST Publication > Article
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE