Searching. Please wait…
1579
37
171
29274
4420
2604
347
391
Abstract: New generations of telecommunications systems will include high-definition 3D video that provides a telepresence feeling. These systems require high-quality depth maps to be generated in a very short time (very low latency, typically about 40 ms). Classical Belief Propagation algorithms (BP) generate high-quality depth maps but they require huge memory bandwidths that limit low-latency implementations of stereo-vision systems with high-definition images. This paper proposes a real-time (latency inferior to 40 ms) high-definition (1280 × 720) stereo matching algorithm using Belief Propagation with good immersive feeling (80 disparity levels). There are two main contributions. The first is an improved BP algorithm with pixel classification that outperforms classical BP while reducing the number of memory accesses. The second is an adaptive message compression technique with a low performance penalty that greatly reduces the memory traffic. The combination of these techniques outperforms classical BP by about 6.0% while reducing the memory traffic by more than 90%.
Authorship: Pérez J.M., Sánchez P.,
Fuente: Pattern Recognition Letters, 2011, 32(16), 2250-2253
Publisher: Elsevier
Publication date: 01/12/2011
No. of pages: 4
Publication type: Article
DOI: 10.1016/j.patrec.2011.06.016
ISSN: 0167-8655,1872-7344
Spanish project: TEC2008- 04107
Publication Url: https://doi.org/10.1016/j.patrec.2011.06.016
Citations in Scopus
Citations in Google Scholar
Other metrics in Scopus
Consult in Scopus
Read publication
JESUS MIGUEL PEREZ LLANO
PABLO PEDRO SANCHEZ ESPESO
Back