Paper Info
Title
From local occlusion cues to global monocular depth estimation
Abstract
In this paper, we propose a system to obtain a depth ordered segmentation of a single image based on low level cues. The algorithm first constructs a hierarchical, region-based image representation of the image using a Binary Partition Tree (BPT). During the building process, T-junction depth cues are detected, along with high convex boundaries. When the BPT is built, a suitable segmentation is found and a global depth ordering is found using a probabilistic framework. Results are compared with state of the art depth ordering and figure/ground labeling systems. The advantage of the proposed approach compared to systems based on a training procedure is the lack of assumptions about the scene content. Moreover, it is shown that the system outperforms previously low-level cue based systems, while offering similar results to a priori trained figure/ground labeling algorithms.
Year
DOI
Venue
2012
10.1109/ICASSP.2012.6288003
Acoustics, Speech and Signal Processing
Keywords
Field
DocType
estimation theory,hidden feature removal,image representation,image segmentation,trees (mathematics),T-junction depth cues,binary partition tree,building process,convex boundaries,depth ordered segmentation,figure/ground labeling systems,global depth ordering,global monocular depth estimation,hierarchical image representation,local occlusion cues,low level cues,probabilistic framework,region-based image representation,T-junctions,binary partition tree,convexity,depth estimation,occlusion
Computer vision,Scale-space segmentation,Pattern recognition,Range segmentation,Segmentation,Computer science,Image texture,Image segmentation,Artificial intelligence,Depth perception,Connected-component labeling,Minimum spanning tree-based segmentation
Conference
ISSN
ISBN
Citations 
1520-6149 E-ISBN : 978-1-4673-0044-5
978-1-4673-0044-5
1
PageRank 
References 
Authors
0.37
6
2
Name
Order
Citations
PageRank
Guillem Palou1141.72
Philippe Salembier260387.65