Abstract | ||
---|---|---|
This paper proposes a system to depth order regions of a frame belonging to a monocular image sequence. For a given frame, regions are ordered according to their relative depth using the previous and following frames. The algorithm estimates occluded and disoccluded pixels belonging to the central frame. Afterwards, a Binary Partition Tree (BPT) is constructed to obtain a hierarchical, region based representation of the image. The final depth partition is obtained by means of energy minimization on the BPT. To achieve a global depth ordering from local occlusion cues, a depth order graph is constructed and used to eliminate contradictory local cues. Results of the system are evaluated and compared with state of the art figure/ground labeling systems on several datasets, showing promising results. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1007/978-3-642-33885-4_52 | ECCV Workshops (3) |
Keywords | Field | DocType |
motion occlusion,relative depth,depth order region,depth estimation,central frame,global depth,depth order graph,final depth partition,binary partition tree,local occlusion cue,contradictory local cue,monocular image sequence | Graph,Computer vision,Data set,Binary partition tree,Parallax,Pattern recognition,Computer science,Monocular image,Artificial intelligence,Pixel,Partition (number theory),Energy minimization | Conference |
Volume | ISSN | Citations |
7585 | 0302-9743 | 1 |
PageRank | References | Authors |
0.35 | 11 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Guillem Palou | 1 | 14 | 0.84 |
Philippe Salembier | 2 | 603 | 87.65 |