Name
Abstract | ||
|---|---|---|
Motion estimation is an important and computationally intensive task in video coding and video analysis. But existent motion estimation algorithms mainly focus on 2-D image plane motion and neglect the motion in depth direction, which we call it depth motion in this paper. There are even few researches on the depth motion, their methods are complex and most of them need binocular images. In this work, visual perception theory is used to estimate the depth motion. A novel depth motion estimate method is proposed base on visual perception theory and it can estimate the depth motion from just monocular video. Experimental results show that our model is simple, effective and corresponds to the human perception. |
Year | DOI | Venue |
|---|---|---|
2007 | 10.1007/978-3-540-69423-6_20 | MMM |
Keywords | Field | DocType |
visual perception,motion estimation,human perception | Structure from motion,Computer vision,Motion field,Quarter-pixel motion,Pattern recognition,Computer science,Motion compensation,Artificial intelligence,Motion estimation,Kinetic depth effect,Optical flow,Motion vector | Conference |
Volume | Issue | ISSN |
4351 LNCS | PART 1 | 0302-9743 |
ISBN | Citations | PageRank |
3-540-69421-8 | 0 | 0.34 |
References | Authors | |
7 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Bing Li | 1 | 4 | 1.79 |
| De Xu | 2 | 88 | 10.45 |
| Songhe Feng | 3 | 218 | 34.57 |
| Fangshi Wang | 4 | 21 | 4.74 |