2001
DOI: 10.1109/34.946992
|View full text |Cite
|
Sign up to set email alerts
|

On active camera control and camera motion recovery with foveate wavelet transform

Abstract: AbstractÐIn this paper, a new variable resolution technique±Foveate Wavelet Transform (FWT) is proposed to represent digital images in an effort to efficiently represent visual data. Compared to existing variable resolution techniques, the strength of the proposed scheme encompasses its linearity preservation, orientation selectivity, and flexibility while supporting interesting behaviors resembling the animate vision system. The linearity preservation of the FWT is due to the fact that only low and/or high-pa… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
8
0

Year Published

2005
2005
2011
2011

Publication Types

Select...
3
3

Relationship

0
6

Authors

Journals

citations
Cited by 10 publications
(8 citation statements)
references
References 23 publications
0
8
0
Order By: Relevance
“…Therefore, the ROI output must be processed for optimal control of the virtual camera. Examples include the 3D virtual cinematographer by He et al [13], 3D animation by Gleicher and Witkin [11], and fovea area view by Wei and Li [32]. Since the work here involves moving a small ROI rectangle inside a large panoramic video, our virtual camera control is equivalent to controlling a moving camera in a 2D image plane.…”
Section: Related Workmentioning
confidence: 99%
“…Therefore, the ROI output must be processed for optimal control of the virtual camera. Examples include the 3D virtual cinematographer by He et al [13], 3D animation by Gleicher and Witkin [11], and fovea area view by Wei and Li [32]. Since the work here involves moving a small ROI rectangle inside a large panoramic video, our virtual camera control is equivalent to controlling a moving camera in a 2D image plane.…”
Section: Related Workmentioning
confidence: 99%
“…One led by Tordoff and Murray (2004) utilizes SFM to recover the target's 3D depth, and the other proposed by Fayman et al (2001) is based on the optical flow of the image sequence. In addition to pure spatial domain processing, the wavelet transform is another promising approach addressed by Wei and Li (2001). Among existing size preserving tracking algorithms, the SFM based algorithm appears to be the most promising primarily due to its superior accuracy and moderate computational complexity.…”
Section: Introductionmentioning
confidence: 99%
“…Although there have been many fruitful results along this line of research, it is very difficult to perform high level vision tasks without active participation from the vision system. It is precisely for this reason that active vision is proposed [3,4] for which efforts are made for computer controlled cameras to actively participate in the visual perception process, much similar to the body and eye movements of human vision systems [5]. Of course, when the camera is operated by a human being as in the case of movie making and even home video making, the lines of reasoning advocated in active vision research can essentially be reversed to form a bridge to connect the conceptual gap between the visual world and the underlying semantic meanings.…”
Section: Introductionmentioning
confidence: 99%